UCI ML Repo: US Census Earnings: 1994: over50k classification
bdanalytics
Date: (Fri) Apr 24, 2015
Introduction:
Data: Source: Training: https://courses.edx.org/c4x/MITx/15.071x_2/asset/census.csv
New:
Time period:
Synopsis:
Based on analysis utilizing <> techniques,
[](.png)
Potential next steps include:
- Organization:
- Categorize by chunk
- Priority criteria:
- Ease of change
- Impacts report
- Cleans innards
- Bug report
- manage.missing.data chunk:
- cleaner way to manage re-splitting of training vs. new entity
- fit.models chunk:
- Prediction accuracy scatter graph:
- Add tiles (raw vs. PCA)
- Use shiny for drop-down of “important” features
Use plot.ly for interactive plots ?
- Change .fit suffix of model metrics to .mdl if it’s data independent (e.g. AIC, Adj.R.Squared - is it truly data independent ?, etc.)
- move model_type parameter to myfit_mdl before indep_vars_vctr (keep all model_* together)
- create a custom model for rpart that has minbucket as a tuning parameter
varImp for randomForest crashes in caret version:6.0.41 -> submit bug report
- Probability handling for multinomials vs. desired binomial outcome
- ROCR currently supports only evaluation of binary classification tasks (version 1.0.7)
extensions toward multiclass classification are scheduled for the next release
- Skip trControl.method=“cv” for dummy classifier ?
- Add custom model to caret for a dummy (baseline) classifier (binomial & multinomial) that generates proba/outcomes which mimics the freq distribution of glb_rsp_var values; Right now glb_dmy_glm_mdl always generates most frequent outcome in training data
glm_dmy_mdl should use the same method as glm_sel_mdl until custom dummy classifer is implemented
- Compare glb_sel_mdl vs. glb_fin_mdl:
- varImp
- Prediction differences (shd be minimal ?)
- Move glb_analytics_diag_plots to mydsutils.R: (+) Easier to debug (-) Too many glb vars used
- Add print(ggplot.petrinet(glb_analytics_pn) + coord_flip()) at the end of every major chunk
- Parameterize glb_analytics_pn
- Move glb_impute_missing_data to mydsutils.R: (-) Too many glb vars used; glb_<>_df reassigned
- Replicate myfit_mdl_classification features in myfit_mdl_regression
- Do non-glm methods handle interaction terms ?
- f-score computation for classifiers should be summation across outcomes (not just the desired one ?)
- Add accuracy computation to glb_dmy_mdl in predict.data.new chunk
- Why does splitting fit.data.training.all chunk into separate chunks add an overhead of ~30 secs ? It’s not rbind b/c other chunks have lower elapsed time. Is it the number of plots ?
- Incorporate code chunks in print_sessionInfo
- Test against
- projects in github.com/bdanalytics
- lectures in jhu-datascience track
Analysis:
rm(list=ls())
set.seed(12345)
options(stringsAsFactors=FALSE)
source("~/Dropbox/datascience/R/mydsutils.R")
source("~/Dropbox/datascience/R/myplot.R")
source("~/Dropbox/datascience/R/mypetrinet.R")
# Gather all package requirements here
#suppressPackageStartupMessages(require())
#packageVersion("snow")
#require(sos); findFn("pinv", maxPages=2, sortby="MaxScore")
# Analysis control global variables
glb_trnng_url <- "https://courses.edx.org/c4x/MITx/15.071x_2/asset/census.csv"
glb_newdt_url <- "<newdt_url>"
glb_is_separate_newent_dataset <- FALSE # or TRUE
glb_split_entity_newent_datasets <- TRUE # or FALSE
glb_split_newdata_method <- "sample" # "condition" or "sample"
glb_split_newdata_condition <- "<col_name> <condition_operator> <value>" # or NULL
glb_split_newdata_size_ratio <- 0.4 # > 0 & < 1
glb_split_sample.seed <- 2000 # or any integer
glb_max_obs <- 1000 # or any integer
glb_is_regression <- FALSE; glb_is_classification <- TRUE; glb_is_binomial <- TRUE
glb_rsp_var_raw <- "over50k"
# for classification, the response variable has to be a factor
glb_rsp_var <- "over50k.fctr"
# if the response factor is based on numbers e.g (0/1 vs. "A"/"B"),
# caret predict(..., type="prob") crashes
glb_map_rsp_raw_to_var <- function(raw) {
relevel(factor(ifelse(raw == " >50K", "Y", "N")), as.factor(c("Y", "N")), ref="N")
#as.factor(paste0("B", raw))
#as.factor(raw)
}
glb_map_rsp_raw_to_var(c(" >50K", " <=50K", " >50K", " <=50K", " <=50K"))## [1] Y N Y N N
## Levels: N Yglb_map_rsp_var_to_raw <- function(var) {
#as.numeric(var) - 1
#as.numeric(var)
c(" <=50K", " >50K")[as.numeric(var)]
}
glb_map_rsp_var_to_raw(glb_map_rsp_raw_to_var(c(" >50K", " <=50K", " >50K", " <=50K", " <=50K")))## [1] " >50K" " <=50K" " >50K" " <=50K" " <=50K"if ((glb_rsp_var != glb_rsp_var_raw) & is.null(glb_map_rsp_raw_to_var))
stop("glb_map_rsp_raw_to_var function expected")
glb_rsp_var_out <- paste0(glb_rsp_var, ".predict.") # model_id is appended later
glb_id_vars <- NULL # or c("<id_var>")
# List transformed vars
glb_exclude_vars_as_features <- c(NULL) # or c("<var_name>")
# List feats that shd be excluded due to known causation by prediction variable
if (glb_rsp_var_raw != glb_rsp_var)
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
glb_rsp_var_raw)
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
c(NULL)) # or c("<col_name>")
# List output vars (useful during testing in console)
# glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features,
# grep(glb_rsp_var_out, names(glb_trnent_df), value=TRUE))
glb_impute_na_data <- FALSE # or TRUE
glb_mice_complete.seed <- 144 # or any integer
# rpart: .rnorm messes with the models badly
# caret creates dummy vars for factor feats which messes up the tuning
# - better to feed as.numeric(<feat>.fctr) to caret
# Regression
if (glb_is_regression)
glb_models_method_vctr <- c("lm", "glm", "rpart", "rf") else
# Classification
if (glb_is_binomial)
glb_models_method_vctr <- c("glm", "rpart", "rf") else
glb_models_method_vctr <- c("rpart", "rf")
glb_models_lst <- list(); glb_models_df <- data.frame()
# Baseline prediction model feature(s)
glb_Baseline_mdl_var <- NULL # or c("<col_name>")
glb_model_metric_terms <- NULL # or matrix(c(
# 0,1,2,3,4,
# 2,0,1,2,3,
# 4,2,0,1,2,
# 6,4,2,0,1,
# 8,6,4,2,0
# ), byrow=TRUE, nrow=5)
glb_model_metric <- NULL # or "<metric_name>"
glb_model_metric_maximize <- NULL # or FALSE (TRUE is not the default for both classification & regression)
glb_model_metric_smmry <- NULL # or function(data, lev=NULL, model=NULL) {
# confusion_mtrx <- t(as.matrix(confusionMatrix(data$pred, data$obs)))
# #print(confusion_mtrx)
# #print(confusion_mtrx * glb_model_metric_terms)
# metric <- sum(confusion_mtrx * glb_model_metric_terms) / nrow(data)
# names(metric) <- glb_model_metric
# return(metric)
# }
glb_tune_models_df <-
rbind(
#data.frame(parameter="cp", min=0.00005, max=0.00005, by=0.000005),
#seq(from=0.01, to=0.01, by=0.01)
#data.frame(parameter="mtry", min=2, max=4, by=1),
data.frame(parameter="dummy", min=2, max=4, by=1)
)
# or NULL
glb_n_cv_folds <- 3 # or NULL
glb_clf_proba_threshold <- NULL # 0.5
# Model selection criteria
if (glb_is_regression)
glb_model_evl_criteria <- c("min.RMSE.OOB", "max.R.sq.OOB", "max.Adj.R.sq.fit")
if (glb_is_classification) {
if (glb_is_binomial)
glb_model_evl_criteria <- c("max.Accuracy.OOB", "max.Kappa.OOB", "min.aic.fit") else
glb_model_evl_criteria <- c("max.Accuracy.OOB", "max.Kappa.OOB")
}
glb_sel_mdl_id <- NULL # or "<model_id_prefix>.<model_method>"
glb_fin_mdl_id <- glb_sel_mdl_id # or "Final"
# Depict process
glb_analytics_pn <- petrinet(name="glb_analytics_pn",
trans_df=data.frame(id=1:6,
name=c("data.training.all","data.new",
"model.selected","model.final",
"data.training.all.prediction","data.new.prediction"),
x=c( -5,-5,-15,-25,-25,-35),
y=c( -5, 5, 0, 0, -5, 5)
),
places_df=data.frame(id=1:4,
name=c("bgn","fit.data.training.all","predict.data.new","end"),
x=c( -0, -20, -30, -40),
y=c( 0, 0, 0, 0),
M0=c( 3, 0, 0, 0)
),
arcs_df=data.frame(
begin=c("bgn","bgn","bgn",
"data.training.all","model.selected","fit.data.training.all",
"fit.data.training.all","model.final",
"data.new","predict.data.new",
"data.training.all.prediction","data.new.prediction"),
end =c("data.training.all","data.new","model.selected",
"fit.data.training.all","fit.data.training.all","model.final",
"data.training.all.prediction","predict.data.new",
"predict.data.new","data.new.prediction",
"end","end")
))
#print(ggplot.petrinet(glb_analytics_pn))
print(ggplot.petrinet(glb_analytics_pn) + coord_flip())## Loading required package: grid
glb_analytics_avl_objs <- NULL
glb_script_tm <- proc.time()
glb_script_df <- data.frame(chunk_label="import_data",
chunk_step_major=1, chunk_step_minor=0,
elapsed=(proc.time() - glb_script_tm)["elapsed"])
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed import_data 1 0 0.002Step 1: import data
glb_entity_df <- myimport_data(url=glb_trnng_url,
comment=ifelse(!glb_is_separate_newent_dataset, "glb_entity_df", "glb_trnent_df"),
force_header=TRUE)## [1] "Reading file ./data/census.csv..."
## [1] "dimensions of data in ./data/census.csv: 31,978 rows x 13 cols"
## age workclass education maritalstatus occupation
## 1 39 State-gov Bachelors Never-married Adm-clerical
## 2 50 Self-emp-not-inc Bachelors Married-civ-spouse Exec-managerial
## 3 38 Private HS-grad Divorced Handlers-cleaners
## 4 53 Private 11th Married-civ-spouse Handlers-cleaners
## 5 28 Private Bachelors Married-civ-spouse Prof-specialty
## 6 37 Private Masters Married-civ-spouse Exec-managerial
## relationship race sex capitalgain capitalloss hoursperweek
## 1 Not-in-family White Male 2174 0 40
## 2 Husband White Male 0 0 13
## 3 Not-in-family White Male 0 0 40
## 4 Husband Black Male 0 0 40
## 5 Wife Black Female 0 0 40
## 6 Wife White Female 0 0 40
## nativecountry over50k
## 1 United-States <=50K
## 2 United-States <=50K
## 3 United-States <=50K
## 4 United-States <=50K
## 5 Cuba <=50K
## 6 United-States <=50K
## age workclass education maritalstatus
## 1105 20 Private Assoc-voc Never-married
## 4872 34 Private HS-grad Never-married
## 10396 33 Private Some-college Divorced
## 16284 70 Self-emp-not-inc Bachelors Married-civ-spouse
## 23270 22 Private HS-grad Never-married
## 31647 19 Private 11th Never-married
## occupation relationship race sex capitalgain
## 1105 Adm-clerical Own-child White Female 0
## 4872 Exec-managerial Other-relative White Male 0
## 10396 Adm-clerical Not-in-family White Female 0
## 16284 Prof-specialty Husband White Male 20051
## 23270 Exec-managerial Own-child White Female 0
## 31647 Sales Own-child Black Female 0
## capitalloss hoursperweek nativecountry over50k
## 1105 0 40 United-States <=50K
## 4872 0 50 United-States <=50K
## 10396 0 40 United-States <=50K
## 16284 0 35 United-States >50K
## 23270 0 40 United-States <=50K
## 31647 0 35 United-States <=50K
## age workclass education maritalstatus
## 31973 22 Private Some-college Never-married
## 31974 27 Private Assoc-acdm Married-civ-spouse
## 31975 40 Private HS-grad Married-civ-spouse
## 31976 58 Private HS-grad Widowed
## 31977 22 Private HS-grad Never-married
## 31978 52 Self-emp-inc HS-grad Married-civ-spouse
## occupation relationship race sex capitalgain
## 31973 Protective-serv Not-in-family White Male 0
## 31974 Tech-support Wife White Female 0
## 31975 Machine-op-inspct Husband White Male 0
## 31976 Adm-clerical Unmarried White Female 0
## 31977 Adm-clerical Own-child White Male 0
## 31978 Exec-managerial Wife White Female 15024
## capitalloss hoursperweek nativecountry over50k
## 31973 0 40 United-States <=50K
## 31974 0 38 United-States <=50K
## 31975 0 40 United-States >50K
## 31976 0 40 United-States <=50K
## 31977 0 20 United-States <=50K
## 31978 0 40 United-States >50K
## 'data.frame': 31978 obs. of 13 variables:
## $ age : int 39 50 38 53 28 37 49 52 31 42 ...
## $ workclass : chr " State-gov" " Self-emp-not-inc" " Private" " Private" ...
## $ education : chr " Bachelors" " Bachelors" " HS-grad" " 11th" ...
## $ maritalstatus: chr " Never-married" " Married-civ-spouse" " Divorced" " Married-civ-spouse" ...
## $ occupation : chr " Adm-clerical" " Exec-managerial" " Handlers-cleaners" " Handlers-cleaners" ...
## $ relationship : chr " Not-in-family" " Husband" " Not-in-family" " Husband" ...
## $ race : chr " White" " White" " White" " Black" ...
## $ sex : chr " Male" " Male" " Male" " Male" ...
## $ capitalgain : int 2174 0 0 0 0 0 0 0 14084 5178 ...
## $ capitalloss : int 0 0 0 0 0 0 0 0 0 0 ...
## $ hoursperweek : int 40 13 40 40 40 40 16 45 50 40 ...
## $ nativecountry: chr " United-States" " United-States" " United-States" " United-States" ...
## $ over50k : chr " <=50K" " <=50K" " <=50K" " <=50K" ...
## - attr(*, "comment")= chr "glb_entity_df"
## NULLif (!glb_is_separate_newent_dataset) {
glb_trnent_df <- glb_entity_df; comment(glb_trnent_df) <- "glb_trnent_df"
} # else glb_entity_df is maintained as is for chunk:inspectORexplore.data
if (glb_is_separate_newent_dataset) {
glb_newent_df <- myimport_data(
url=glb_newdt_url,
comment="glb_newent_df", force_header=TRUE)
# To make plots / stats / checks easier in chunk:inspectORexplore.data
glb_entity_df <- rbind(glb_trnent_df, glb_newent_df); comment(glb_entity_df) <- "glb_entity_df"
} else {
if (!glb_split_entity_newent_datasets) {
stop("Not implemented yet")
glb_newent_df <- glb_trnent_df[sample(1:nrow(glb_trnent_df),
max(2, nrow(glb_trnent_df) / 1000)),]
} else if (glb_split_newdata_method == "condition") {
glb_newent_df <- do.call("subset",
list(glb_trnent_df, parse(text=glb_split_newdata_condition)))
glb_trnent_df <- do.call("subset",
list(glb_trnent_df, parse(text=paste0("!(",
glb_split_newdata_condition,
")"))))
} else if (glb_split_newdata_method == "sample") {
require(caTools)
set.seed(glb_split_sample.seed)
split <- sample.split(glb_trnent_df[, glb_rsp_var_raw],
SplitRatio=(1-glb_split_newdata_size_ratio))
glb_newent_df <- glb_trnent_df[!split, ]
glb_trnent_df <- glb_trnent_df[split ,]
} else stop("glb_split_newdata_method should be %in% c('condition', 'sample')")
comment(glb_newent_df) <- "glb_newent_df"
myprint_df(glb_newent_df)
str(glb_newent_df)
if (glb_split_entity_newent_datasets) {
myprint_df(glb_trnent_df)
str(glb_trnent_df)
}
} ## Loading required package: caTools## age workclass education maritalstatus
## 2 50 Self-emp-not-inc Bachelors Married-civ-spouse
## 5 28 Private Bachelors Married-civ-spouse
## 7 49 Private 9th Married-spouse-absent
## 8 52 Self-emp-not-inc HS-grad Married-civ-spouse
## 11 37 Private Some-college Married-civ-spouse
## 12 30 State-gov Bachelors Married-civ-spouse
## occupation relationship race sex capitalgain
## 2 Exec-managerial Husband White Male 0
## 5 Prof-specialty Wife Black Female 0
## 7 Other-service Not-in-family Black Female 0
## 8 Exec-managerial Husband White Male 0
## 11 Exec-managerial Husband Black Male 0
## 12 Prof-specialty Husband Asian-Pac-Islander Male 0
## capitalloss hoursperweek nativecountry over50k
## 2 0 13 United-States <=50K
## 5 0 40 Cuba <=50K
## 7 0 16 Jamaica <=50K
## 8 0 45 United-States >50K
## 11 0 80 United-States >50K
## 12 0 40 India >50K
## age workclass education maritalstatus
## 19 43 Self-emp-not-inc Masters Divorced
## 13546 44 Private Assoc-acdm Divorced
## 16559 49 Private Assoc-voc Married-civ-spouse
## 23733 22 Private HS-grad Never-married
## 29170 34 Private HS-grad Divorced
## 30522 26 Private Bachelors Never-married
## occupation relationship race sex
## 19 Exec-managerial Unmarried White Female
## 13546 Machine-op-inspct Own-child White Female
## 16559 Adm-clerical Husband White Male
## 23733 Adm-clerical Own-child White Female
## 29170 Transport-moving Not-in-family White Male
## 30522 Exec-managerial Unmarried Asian-Pac-Islander Male
## capitalgain capitalloss hoursperweek nativecountry over50k
## 19 0 0 45 United-States >50K
## 13546 0 0 45 United-States <=50K
## 16559 0 0 40 United-States <=50K
## 23733 0 0 40 United-States <=50K
## 29170 0 0 50 United-States <=50K
## 30522 0 0 60 Vietnam <=50K
## age workclass education maritalstatus
## 31966 65 Self-emp-not-inc Prof-school Never-married
## 31968 43 Self-emp-not-inc Some-college Married-civ-spouse
## 31972 53 Private Masters Married-civ-spouse
## 31974 27 Private Assoc-acdm Married-civ-spouse
## 31975 40 Private HS-grad Married-civ-spouse
## 31976 58 Private HS-grad Widowed
## occupation relationship race sex capitalgain
## 31966 Prof-specialty Not-in-family White Male 1086
## 31968 Craft-repair Husband White Male 0
## 31972 Exec-managerial Husband White Male 0
## 31974 Tech-support Wife White Female 0
## 31975 Machine-op-inspct Husband White Male 0
## 31976 Adm-clerical Unmarried White Female 0
## capitalloss hoursperweek nativecountry over50k
## 31966 0 60 United-States <=50K
## 31968 0 50 United-States <=50K
## 31972 0 40 United-States >50K
## 31974 0 38 United-States <=50K
## 31975 0 40 United-States >50K
## 31976 0 40 United-States <=50K
## 'data.frame': 12791 obs. of 13 variables:
## $ age : int 50 28 49 52 37 30 23 32 34 43 ...
## $ workclass : chr " Self-emp-not-inc" " Private" " Private" " Self-emp-not-inc" ...
## $ education : chr " Bachelors" " Bachelors" " 9th" " HS-grad" ...
## $ maritalstatus: chr " Married-civ-spouse" " Married-civ-spouse" " Married-spouse-absent" " Married-civ-spouse" ...
## $ occupation : chr " Exec-managerial" " Prof-specialty" " Other-service" " Exec-managerial" ...
## $ relationship : chr " Husband" " Wife" " Not-in-family" " Husband" ...
## $ race : chr " White" " Black" " Black" " White" ...
## $ sex : chr " Male" " Female" " Female" " Male" ...
## $ capitalgain : int 0 0 0 0 0 0 0 0 0 0 ...
## $ capitalloss : int 0 0 0 0 0 0 0 0 0 0 ...
## $ hoursperweek : int 13 40 16 45 80 40 30 50 45 45 ...
## $ nativecountry: chr " United-States" " Cuba" " Jamaica" " United-States" ...
## $ over50k : chr " <=50K" " <=50K" " <=50K" " >50K" ...
## - attr(*, "comment")= chr "glb_newent_df"
## age workclass education maritalstatus occupation
## 1 39 State-gov Bachelors Never-married Adm-clerical
## 3 38 Private HS-grad Divorced Handlers-cleaners
## 4 53 Private 11th Married-civ-spouse Handlers-cleaners
## 6 37 Private Masters Married-civ-spouse Exec-managerial
## 9 31 Private Masters Never-married Prof-specialty
## 10 42 Private Bachelors Married-civ-spouse Exec-managerial
## relationship race sex capitalgain capitalloss hoursperweek
## 1 Not-in-family White Male 2174 0 40
## 3 Not-in-family White Male 0 0 40
## 4 Husband Black Male 0 0 40
## 6 Wife White Female 0 0 40
## 9 Not-in-family White Female 14084 0 50
## 10 Husband White Male 5178 0 40
## nativecountry over50k
## 1 United-States <=50K
## 3 United-States <=50K
## 4 United-States <=50K
## 6 United-States <=50K
## 9 United-States >50K
## 10 United-States >50K
## age workclass education maritalstatus
## 3944 19 ? Some-college Never-married
## 4463 38 Self-emp-not-inc Some-college Married-civ-spouse
## 8872 30 Private HS-grad Married-civ-spouse
## 15414 28 Private HS-grad Never-married
## 17475 67 Private 1st-4th Widowed
## 30989 42 State-gov Some-college Divorced
## occupation relationship race sex capitalgain
## 3944 ? Own-child White Female 0
## 4463 Machine-op-inspct Husband White Male 7688
## 8872 Other-service Husband Black Male 0
## 15414 Machine-op-inspct Not-in-family White Male 0
## 17475 Machine-op-inspct Not-in-family White Female 2062
## 30989 Adm-clerical Unmarried Black Female 0
## capitalloss hoursperweek nativecountry over50k
## 3944 0 45 United-States <=50K
## 4463 0 40 United-States >50K
## 8872 0 25 United-States <=50K
## 15414 0 50 United-States <=50K
## 17475 0 34 Ecuador <=50K
## 30989 0 40 United-States <=50K
## age workclass education maritalstatus
## 31969 32 Private 10th Married-civ-spouse
## 31970 43 Private Assoc-voc Married-civ-spouse
## 31971 32 Private Masters Never-married
## 31973 22 Private Some-college Never-married
## 31977 22 Private HS-grad Never-married
## 31978 52 Self-emp-inc HS-grad Married-civ-spouse
## occupation relationship race sex
## 31969 Handlers-cleaners Husband Amer-Indian-Eskimo Male
## 31970 Sales Husband White Male
## 31971 Tech-support Not-in-family Asian-Pac-Islander Male
## 31973 Protective-serv Not-in-family White Male
## 31977 Adm-clerical Own-child White Male
## 31978 Exec-managerial Wife White Female
## capitalgain capitalloss hoursperweek nativecountry over50k
## 31969 0 0 40 United-States <=50K
## 31970 0 0 45 United-States <=50K
## 31971 0 0 11 Taiwan <=50K
## 31973 0 0 40 United-States <=50K
## 31977 0 0 20 United-States <=50K
## 31978 15024 0 40 United-States >50K
## 'data.frame': 19187 obs. of 13 variables:
## $ age : int 39 38 53 37 31 42 25 32 38 54 ...
## $ workclass : chr " State-gov" " Private" " Private" " Private" ...
## $ education : chr " Bachelors" " HS-grad" " 11th" " Masters" ...
## $ maritalstatus: chr " Never-married" " Divorced" " Married-civ-spouse" " Married-civ-spouse" ...
## $ occupation : chr " Adm-clerical" " Handlers-cleaners" " Handlers-cleaners" " Exec-managerial" ...
## $ relationship : chr " Not-in-family" " Not-in-family" " Husband" " Wife" ...
## $ race : chr " White" " White" " Black" " White" ...
## $ sex : chr " Male" " Male" " Male" " Female" ...
## $ capitalgain : int 2174 0 0 0 14084 5178 0 0 0 0 ...
## $ capitalloss : int 0 0 0 0 0 0 0 0 0 0 ...
## $ hoursperweek : int 40 40 40 40 50 40 35 40 50 20 ...
## $ nativecountry: chr " United-States" " United-States" " United-States" " United-States" ...
## $ over50k : chr " <=50K" " <=50K" " <=50K" " <=50K" ...
## - attr(*, "comment")= chr "glb_trnent_df"if (!is.null(glb_max_obs)) {
if (nrow(glb_trnent_df) > glb_max_obs) {
warning("glb_trnent_df restricted to glb_max_obs: ", format(glb_max_obs, big.mark=","))
org_entity_df <- glb_trnent_df
glb_trnent_df <- org_entity_df[split <-
sample.split(org_entity_df[, glb_rsp_var_raw], SplitRatio=glb_max_obs), ]
org_entity_df <- NULL
}
if (nrow(glb_newent_df) > glb_max_obs) {
warning("glb_newent_df restricted to glb_max_obs: ", format(glb_max_obs, big.mark=","))
org_newent_df <- glb_newent_df
glb_newent_df <- org_newent_df[split <-
sample.split(org_newent_df[, glb_rsp_var_raw], SplitRatio=glb_max_obs), ]
org_newent_df <- NULL
}
}## Warning: glb_trnent_df restricted to glb_max_obs: 1,000## Warning: glb_newent_df restricted to glb_max_obs: 1,000glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="cleanse_data",
chunk_step_major=max(glb_script_df$chunk_step_major)+1,
chunk_step_minor=0,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed import_data 1 0 0.002
## elapsed1 cleanse_data 2 0 1.654Step 2: cleanse data
glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="inspectORexplore.data",
chunk_step_major=max(glb_script_df$chunk_step_major),
chunk_step_minor=1,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed1 cleanse_data 2 0 1.654
## elapsed2 inspectORexplore.data 2 1 1.692Step 2.1: inspect/explore data
#print(str(glb_trnent_df))
#View(glb_trnent_df)
# List info gathered for various columns
# <col_name>: <description>; <notes>
# age = the age of the individual in years
# workclass = the classification of the individual's working status (does the person work for the federal government, work for the local government, work without pay, and so on)
# education = the level of education of the individual (e.g., 5th-6th grade, high school graduate, PhD, so on)
# maritalstatus = the marital status of the individual
# occupation = the type of work the individual does (e.g., administrative/clerical work, farming/fishing, sales and so on)
# relationship = relationship of individual to his/her household
# race = the individual's race
# sex = the individual's sex
# capitalgain = the capital gains of the individual in 1994 (from selling an asset such as a stock or bond for more than the original purchase price)
# capitalloss = the capital losses of the individual in 1994 (from selling an asset such as a stock or bond for less than the original purchase price)
# hoursperweek = the number of hours the individual works per week
# nativecountry = the native country of the individual
# over50k = whether or not the individual earned more than $50,000 in 1994
# Create new features that help diagnostics
# Create factors of string variables
str_vars <- sapply(1:length(names(glb_trnent_df)),
function(col) ifelse(class(glb_trnent_df[, names(glb_trnent_df)[col]]) == "character",
names(glb_trnent_df)[col], ""))
if (length(str_vars <- setdiff(str_vars[str_vars != ""],
glb_exclude_vars_as_features)) > 0) {
warning("Creating factors of string variables:", paste0(str_vars, collapse=", "))
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features, str_vars)
for (var in str_vars) {
glb_entity_df[, paste0(var, ".fctr")] <- factor(glb_entity_df[, var],
as.factor(unique(glb_entity_df[, var])))
glb_trnent_df[, paste0(var, ".fctr")] <- factor(glb_trnent_df[, var],
as.factor(unique(glb_entity_df[, var])))
glb_newent_df[, paste0(var, ".fctr")] <- factor(glb_newent_df[, var],
as.factor(unique(glb_entity_df[, var])))
}
}## Warning: Creating factors of string variables:workclass, education,
## maritalstatus, occupation, relationship, race, sex, nativecountry# Convert factors to dummy variables
# Build splines require(splines); bsBasis <- bs(training$age, df=3)
add_new_diag_feats <- function(obs_df, ref_df=glb_entity_df) {
require(plyr)
obs_df <- mutate(obs_df,
# <col_name>.NA=is.na(<col_name>),
# <col_name>.fctr=factor(<col_name>,
# as.factor(union(obs_df$<col_name>, obs_twin_df$<col_name>))),
# <col_name>.fctr=relevel(factor(<col_name>,
# as.factor(union(obs_df$<col_name>, obs_twin_df$<col_name>))),
# "<ref_val>"),
# <col2_name>.fctr=relevel(factor(ifelse(<col1_name> == <val>, "<oth_val>", "<ref_val>")),
# as.factor(c("R", "<ref_val>")),
# ref="<ref_val>"),
# This doesn't work - use sapply instead
# <col_name>.fctr_num=grep(<col_name>, levels(<col_name>.fctr)),
#
# Date.my=as.Date(strptime(Date, "%m/%d/%y %H:%M")),
# Year=year(Date.my),
# Month=months(Date.my),
# Weekday=weekdays(Date.my)
# <col_name>.log=log(<col.name>),
# <col_name>=<table>[as.character(<col2_name>)],
# <col_name>=as.numeric(<col2_name>),
.rnorm=rnorm(n=nrow(obs_df))
)
# If levels of a factor are different across obs_df & glb_newent_df; predict.glm fails
# Transformations not handled by mutate
# obs_df$<col_name>.fctr.num <- sapply(1:nrow(obs_df),
# function(row_ix) grep(obs_df[row_ix, "<col_name>"],
# levels(obs_df[row_ix, "<col_name>.fctr"])))
print(summary(obs_df))
print(sapply(names(obs_df), function(col) sum(is.na(obs_df[, col]))))
return(obs_df)
}
glb_entity_df <- add_new_diag_feats(glb_entity_df)## Loading required package: plyr## age workclass education maritalstatus
## Min. :17.00 Length:31978 Length:31978 Length:31978
## 1st Qu.:28.00 Class :character Class :character Class :character
## Median :37.00 Mode :character Mode :character Mode :character
## Mean :38.58
## 3rd Qu.:48.00
## Max. :90.00
##
## occupation relationship race
## Length:31978 Length:31978 Length:31978
## Class :character Class :character Class :character
## Mode :character Mode :character Mode :character
##
##
##
##
## sex capitalgain capitalloss hoursperweek
## Length:31978 Min. : 0 Min. : 0.00 Min. : 1.00
## Class :character 1st Qu.: 0 1st Qu.: 0.00 1st Qu.:40.00
## Mode :character Median : 0 Median : 0.00 Median :40.00
## Mean : 1064 Mean : 86.74 Mean :40.42
## 3rd Qu.: 0 3rd Qu.: 0.00 3rd Qu.:45.00
## Max. :99999 Max. :4356.00 Max. :99.00
##
## nativecountry over50k workclass.fctr
## Length:31978 Length:31978 Private :22286
## Class :character Class :character Self-emp-not-inc: 2499
## Mode :character Mode :character Local-gov : 2067
## ? : 1809
## State-gov : 1279
## Self-emp-inc : 1074
## (Other) : 964
## education.fctr maritalstatus.fctr
## HS-grad :10368 Never-married :10488
## Some-college: 7187 Married-civ-spouse :14692
## Bachelors : 5210 Divorced : 4394
## Masters : 1674 Married-spouse-absent: 397
## Assoc-voc : 1366 Separated : 1005
## 11th : 1167 Married-AF-spouse : 23
## (Other) : 5006 Widowed : 979
## occupation.fctr relationship.fctr
## Prof-specialty :4038 Not-in-family : 8156
## Craft-repair :4030 Husband :12947
## Exec-managerial:3992 Wife : 1534
## Adm-clerical :3721 Own-child : 5005
## Sales :3584 Unmarried : 3384
## Other-service :3212 Other-relative: 952
## (Other) :9401
## race.fctr sex.fctr nativecountry.fctr
## White :27430 Male :21370 United-States:29170
## Black : 3028 Female:10608 Mexico : 643
## Asian-Pac-Islander: 956 Philippines : 198
## Amer-Indian-Eskimo: 311 Germany : 137
## Other : 253 Canada : 121
## Puerto-Rico : 114
## (Other) : 1595
## .rnorm
## Min. :-4.450396
## 1st Qu.:-0.685706
## Median :-0.000782
## Mean :-0.006829
## 3rd Qu.: 0.669782
## Max. : 4.108640
##
## age workclass education
## 0 0 0
## maritalstatus occupation relationship
## 0 0 0
## race sex capitalgain
## 0 0 0
## capitalloss hoursperweek nativecountry
## 0 0 0
## over50k workclass.fctr education.fctr
## 0 0 0
## maritalstatus.fctr occupation.fctr relationship.fctr
## 0 0 0
## race.fctr sex.fctr nativecountry.fctr
## 0 0 0
## .rnorm
## 0glb_trnent_df <- add_new_diag_feats(glb_trnent_df)## age workclass education maritalstatus
## Min. :17.00 Length:1000 Length:1000 Length:1000
## 1st Qu.:28.00 Class :character Class :character Class :character
## Median :37.00 Mode :character Mode :character Mode :character
## Mean :38.88
## 3rd Qu.:47.00
## Max. :90.00
##
## occupation relationship race
## Length:1000 Length:1000 Length:1000
## Class :character Class :character Class :character
## Mode :character Mode :character Mode :character
##
##
##
##
## sex capitalgain capitalloss hoursperweek
## Length:1000 Min. : 0 Min. : 0.00 Min. : 1.0
## Class :character 1st Qu.: 0 1st Qu.: 0.00 1st Qu.:40.0
## Mode :character Median : 0 Median : 0.00 Median :40.0
## Mean : 1107 Mean : 91.74 Mean :40.7
## 3rd Qu.: 0 3rd Qu.: 0.00 3rd Qu.:45.0
## Max. :99999 Max. :2457.00 Max. :99.0
##
## nativecountry over50k workclass.fctr
## Length:1000 Length:1000 Private :685
## Class :character Class :character Self-emp-not-inc: 77
## Mode :character Mode :character Local-gov : 77
## ? : 52
## Self-emp-inc : 42
## State-gov : 36
## (Other) : 31
## education.fctr maritalstatus.fctr
## HS-grad :326 Never-married :314
## Some-college:251 Married-civ-spouse :461
## Bachelors :155 Divorced :148
## Masters : 60 Married-spouse-absent: 10
## Assoc-voc : 46 Separated : 36
## 10th : 37 Married-AF-spouse : 1
## (Other) :125 Widowed : 30
## occupation.fctr relationship.fctr race.fctr
## Exec-managerial:129 Not-in-family :245 White :862
## Prof-specialty :129 Husband :408 Black : 88
## Craft-repair :125 Wife : 52 Asian-Pac-Islander: 30
## Adm-clerical :122 Own-child :159 Amer-Indian-Eskimo: 9
## Sales :112 Unmarried :114 Other : 11
## Other-service :103 Other-relative: 22
## (Other) :280
## sex.fctr nativecountry.fctr .rnorm
## Male :664 United-States:910 Min. :-3.04264
## Female:336 Mexico : 21 1st Qu.:-0.59901
## India : 6 Median : 0.02307
## El-Salvador : 6 Mean : 0.02115
## England : 5 3rd Qu.: 0.68508
## Germany : 5 Max. : 3.63423
## (Other) : 47
## age workclass education
## 0 0 0
## maritalstatus occupation relationship
## 0 0 0
## race sex capitalgain
## 0 0 0
## capitalloss hoursperweek nativecountry
## 0 0 0
## over50k workclass.fctr education.fctr
## 0 0 0
## maritalstatus.fctr occupation.fctr relationship.fctr
## 0 0 0
## race.fctr sex.fctr nativecountry.fctr
## 0 0 0
## .rnorm
## 0glb_newent_df <- add_new_diag_feats(glb_newent_df)## age workclass education maritalstatus
## Min. :17.00 Length:1000 Length:1000 Length:1000
## 1st Qu.:27.00 Class :character Class :character Class :character
## Median :37.00 Mode :character Mode :character Mode :character
## Mean :38.56
## 3rd Qu.:47.25
## Max. :82.00
##
## occupation relationship race
## Length:1000 Length:1000 Length:1000
## Class :character Class :character Class :character
## Mode :character Mode :character Mode :character
##
##
##
##
## sex capitalgain capitalloss hoursperweek
## Length:1000 Min. : 0 Min. : 0.00 Min. : 3.00
## Class :character 1st Qu.: 0 1st Qu.: 0.00 1st Qu.:40.00
## Mode :character Median : 0 Median : 0.00 Median :40.00
## Mean : 1183 Mean : 97.81 Mean :39.57
## 3rd Qu.: 0 3rd Qu.: 0.00 3rd Qu.:45.00
## Max. :99999 Max. :2392.00 Max. :99.00
##
## nativecountry over50k workclass.fctr
## Length:1000 Length:1000 Private :698
## Class :character Class :character Local-gov : 73
## Mode :character Mode :character ? : 69
## Self-emp-not-inc: 67
## Federal-gov : 34
## State-gov : 33
## (Other) : 26
## education.fctr maritalstatus.fctr
## HS-grad :323 Never-married :318
## Some-college:222 Married-civ-spouse :467
## Bachelors :159 Divorced :141
## Masters : 68 Married-spouse-absent: 13
## 11th : 44 Separated : 27
## Assoc-voc : 39 Married-AF-spouse : 1
## (Other) :145 Widowed : 33
## occupation.fctr relationship.fctr race.fctr
## Exec-managerial:137 Not-in-family :254 White :865
## Prof-specialty :130 Husband :407 Black : 95
## Sales :121 Wife : 51 Asian-Pac-Islander: 29
## Craft-repair :120 Own-child :148 Amer-Indian-Eskimo: 5
## Adm-clerical :116 Unmarried :112 Other : 6
## Other-service : 93 Other-relative: 28
## (Other) :283
## sex.fctr nativecountry.fctr .rnorm
## Male :647 United-States:905 Min. :-3.11383
## Female:353 Mexico : 19 1st Qu.:-0.73504
## Germany : 8 Median :-0.07260
## Philippines : 7 Mean :-0.05372
## Iran : 5 3rd Qu.: 0.61448
## Puerto-Rico : 4 Max. : 2.85309
## (Other) : 52
## age workclass education
## 0 0 0
## maritalstatus occupation relationship
## 0 0 0
## race sex capitalgain
## 0 0 0
## capitalloss hoursperweek nativecountry
## 0 0 0
## over50k workclass.fctr education.fctr
## 0 0 0
## maritalstatus.fctr occupation.fctr relationship.fctr
## 0 0 0
## race.fctr sex.fctr nativecountry.fctr
## 0 0 0
## .rnorm
## 0# Histogram of predictor in glb_trnent_df & glb_newent_df
plot_df <- rbind(cbind(glb_trnent_df[, glb_rsp_var_raw, FALSE], data.frame(.data="Training")),
cbind(glb_trnent_df[, glb_rsp_var_raw, FALSE], data.frame(.data="New")))
print(myplot_histogram(plot_df, glb_rsp_var_raw) + facet_wrap(~ .data))## Warning in mean.default(sort(x, partial = half + 0L:1L)[half + 0L:1L]):
## argument is not numeric or logical: returning NA## Warning: Removed 1 rows containing missing values (geom_segment).## Warning: Removed 1 rows containing missing values (geom_segment).
if (glb_is_classification) {
xtab_df <- mycreate_xtab(plot_df, c(".data", glb_rsp_var_raw))
rownames(xtab_df) <- xtab_df$.data
xtab_df <- subset(xtab_df, select=-.data)
print(xtab_df / rowSums(xtab_df))
} ## Loading required package: reshape2## over50k. <=50K over50k. >50K
## New 0.759 0.241
## Training 0.759 0.241# Check for duplicates in glb_id_vars
if (length(glb_id_vars) > 0) {
id_vars_dups_df <- subset(id_vars_df <-
mycreate_tbl_df(glb_entity_df[, glb_id_vars, FALSE], glb_id_vars),
.freq > 1)
if (nrow(id_vars_dups_df) > 0) {
warning("Duplicates found in glb_id_vars data:", nrow(id_vars_dups_df))
myprint_df(id_vars_dups_df)
} else {
# glb_id_vars are unique across obs in both glb_<>_df
glb_exclude_vars_as_features <- union(glb_exclude_vars_as_features, glb_id_vars)
}
}
#pairs(subset(glb_trnent_df, select=-c(col_symbol)))
# Check for glb_newent_df & glb_trnent_df features range mismatches
# Other diagnostics:
# print(subset(glb_trnent_df, <col1_name> == max(glb_trnent_df$<col1_name>, na.rm=TRUE) &
# <col2_name> <= mean(glb_trnent_df$<col1_name>, na.rm=TRUE)))
# print(glb_trnent_df[which.max(glb_trnent_df$<col_name>),])
# print(<col_name>_freq_glb_trnent_df <- mycreate_tbl_df(glb_trnent_df, "<col_name>"))
# print(which.min(table(glb_trnent_df$<col_name>)))
# print(which.max(table(glb_trnent_df$<col_name>)))
# print(which.max(table(glb_trnent_df$<col1_name>, glb_trnent_df$<col2_name>)[, 2]))
# print(table(glb_trnent_df$<col1_name>, glb_trnent_df$<col2_name>))
# print(table(is.na(glb_trnent_df$<col1_name>), glb_trnent_df$<col2_name>))
# print(table(sign(glb_trnent_df$<col1_name>), glb_trnent_df$<col2_name>))
# print(mycreate_xtab(glb_trnent_df, <col1_name>))
# print(mycreate_xtab(glb_trnent_df, c(<col1_name>, <col2_name>)))
# print(<col1_name>_<col2_name>_xtab_glb_trnent_df <-
# mycreate_xtab(glb_trnent_df, c("<col1_name>", "<col2_name>")))
# <col1_name>_<col2_name>_xtab_glb_trnent_df[is.na(<col1_name>_<col2_name>_xtab_glb_trnent_df)] <- 0
# print(<col1_name>_<col2_name>_xtab_glb_trnent_df <-
# mutate(<col1_name>_<col2_name>_xtab_glb_trnent_df,
# <col3_name>=(<col1_name> * 1.0) / (<col1_name> + <col2_name>)))
# print(<col2_name>_min_entity_arr <-
# sort(tapply(glb_trnent_df$<col1_name>, glb_trnent_df$<col2_name>, min, na.rm=TRUE)))
# print(<col1_name>_na_by_<col2_name>_arr <-
# sort(tapply(glb_trnent_df$<col1_name>.NA, glb_trnent_df$<col2_name>, mean, na.rm=TRUE)))
# Other plots:
# print(myplot_box(df=glb_trnent_df, ycol_names="<col1_name>"))
# print(myplot_box(df=glb_trnent_df, ycol_names="<col1_name>", xcol_name="<col2_name>"))
# print(myplot_line(subset(glb_trnent_df, Symbol %in% c("KO", "PG")),
# "Date.my", "StockPrice", facet_row_colnames="Symbol") +
# geom_vline(xintercept=as.numeric(as.Date("2003-03-01"))) +
# geom_vline(xintercept=as.numeric(as.Date("1983-01-01")))
# )
# print(myplot_scatter(glb_trnent_df, "<col1_name>", "<col2_name>", smooth=TRUE))
# print(myplot_scatter(glb_entity_df, "<col1_name>", "<col2_name>", colorcol_name="<Pred.fctr>") +
# geom_point(data=subset(glb_entity_df, <condition>),
# mapping=aes(x=<x_var>, y=<y_var>), color="red", shape=4, size=5))
glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="manage_missing_data",
chunk_step_major=max(glb_script_df$chunk_step_major),
chunk_step_minor=glb_script_df[nrow(glb_script_df), "chunk_step_minor"]+1,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed2 inspectORexplore.data 2 1 1.692
## elapsed3 manage_missing_data 2 2 2.860Step 2.2: manage missing data
# print(sapply(names(glb_trnent_df), function(col) sum(is.na(glb_trnent_df[, col]))))
# print(sapply(names(glb_newent_df), function(col) sum(is.na(glb_newent_df[, col]))))
# glb_trnent_df <- na.omit(glb_trnent_df)
# glb_newent_df <- na.omit(glb_newent_df)
# df[is.na(df)] <- 0
# Not refactored into mydsutils.R since glb_*_df might be reassigned
glb_impute_missing_data <- function(entity_df, newent_df) {
if (!glb_is_separate_newent_dataset) {
# Combine entity & newent
union_df <- rbind(mutate(entity_df, .src = "entity"),
mutate(newent_df, .src = "newent"))
union_imputed_df <- union_df[, setdiff(setdiff(names(entity_df),
glb_rsp_var),
glb_exclude_vars_as_features)]
print(summary(union_imputed_df))
require(mice)
set.seed(glb_mice_complete.seed)
union_imputed_df <- complete(mice(union_imputed_df))
print(summary(union_imputed_df))
union_df[, names(union_imputed_df)] <- union_imputed_df[, names(union_imputed_df)]
print(summary(union_df))
# union_df$.rownames <- rownames(union_df)
# union_df <- orderBy(~.rownames, union_df)
#
# imp_entity_df <- myimport_data(
# url="<imputed_trnng_url>",
# comment="imp_entity_df", force_header=TRUE, print_diagn=TRUE)
# print(all.equal(subset(union_df, select=-c(.src, .rownames, .rnorm)),
# imp_entity_df))
# Partition again
glb_trnent_df <<- subset(union_df, .src == "entity", select=-c(.src, .rownames))
comment(glb_trnent_df) <- "entity_df"
glb_newent_df <<- subset(union_df, .src == "newent", select=-c(.src, .rownames))
comment(glb_newent_df) <- "newent_df"
# Generate summaries
print(summary(entity_df))
print(sapply(names(entity_df), function(col) sum(is.na(entity_df[, col]))))
print(summary(newent_df))
print(sapply(names(newent_df), function(col) sum(is.na(newent_df[, col]))))
} else stop("Not implemented yet")
}
if (glb_impute_na_data) {
if ((sum(sapply(names(glb_trnent_df),
function(col) sum(is.na(glb_trnent_df[, col])))) > 0) |
(sum(sapply(names(glb_newent_df),
function(col) sum(is.na(glb_newent_df[, col])))) > 0))
glb_impute_missing_data(glb_trnent_df, glb_newent_df)
}
glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="encode_retype_data",
chunk_step_major=max(glb_script_df$chunk_step_major),
chunk_step_minor=glb_script_df[nrow(glb_script_df), "chunk_step_minor"]+1,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed3 manage_missing_data 2 2 2.860
## elapsed4 encode_retype_data 2 3 3.141Step 2.3: encode/retype data
# map_<col_name>_df <- myimport_data(
# url="<map_url>",
# comment="map_<col_name>_df", print_diagn=TRUE)
# map_<col_name>_df <- read.csv(paste0(getwd(), "/data/<file_name>.csv"), strip.white=TRUE)
# glb_trnent_df <- mymap_codes(glb_trnent_df, "<from_col_name>", "<to_col_name>",
# map_<to_col_name>_df, map_join_col_name="<map_join_col_name>",
# map_tgt_col_name="<to_col_name>")
# glb_newent_df <- mymap_codes(glb_newent_df, "<from_col_name>", "<to_col_name>",
# map_<to_col_name>_df, map_join_col_name="<map_join_col_name>",
# map_tgt_col_name="<to_col_name>")
# glb_trnent_df$<col_name>.fctr <- factor(glb_trnent_df$<col_name>,
# as.factor(union(glb_trnent_df$<col_name>, glb_newent_df$<col_name>)))
# glb_newent_df$<col_name>.fctr <- factor(glb_newent_df$<col_name>,
# as.factor(union(glb_trnent_df$<col_name>, glb_newent_df$<col_name>)))
if (!is.null(glb_map_rsp_raw_to_var)) {
glb_entity_df[, glb_rsp_var] <-
glb_map_rsp_raw_to_var(glb_entity_df[, glb_rsp_var_raw])
mycheck_map_results(mapd_df=glb_entity_df,
from_col_name=glb_rsp_var_raw, to_col_name=glb_rsp_var)
glb_trnent_df[, glb_rsp_var] <-
glb_map_rsp_raw_to_var(glb_trnent_df[, glb_rsp_var_raw])
mycheck_map_results(mapd_df=glb_trnent_df,
from_col_name=glb_rsp_var_raw, to_col_name=glb_rsp_var)
glb_newent_df[, glb_rsp_var] <-
glb_map_rsp_raw_to_var(glb_newent_df[, glb_rsp_var_raw])
mycheck_map_results(mapd_df=glb_newent_df,
from_col_name=glb_rsp_var_raw, to_col_name=glb_rsp_var)
}## Loading required package: sqldf
## Loading required package: gsubfn
## Loading required package: proto
## Loading required package: RSQLite
## Loading required package: DBI
## Loading required package: tcltk## over50k over50k.fctr .n
## 1 <=50K N 24283
## 2 >50K Y 7695
## over50k over50k.fctr .n
## 1 <=50K N 759
## 2 >50K Y 241
## over50k over50k.fctr .n
## 1 <=50K N 759
## 2 >50K Y 241
glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="extract_features",
chunk_step_major=max(glb_script_df$chunk_step_major)+1,
chunk_step_minor=0,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed4 encode_retype_data 2 3 3.141
## elapsed5 extract_features 3 0 6.361Step 3: extract features
# Create new features that help prediction
# <col_name>.lag.2 <- lag(zoo(glb_trnent_df$<col_name>), -2, na.pad=TRUE)
# glb_trnent_df[, "<col_name>.lag.2"] <- coredata(<col_name>.lag.2)
# <col_name>.lag.2 <- lag(zoo(glb_newent_df$<col_name>), -2, na.pad=TRUE)
# glb_newent_df[, "<col_name>.lag.2"] <- coredata(<col_name>.lag.2)
#
# glb_newent_df[1, "<col_name>.lag.2"] <- glb_trnent_df[nrow(glb_trnent_df) - 1,
# "<col_name>"]
# glb_newent_df[2, "<col_name>.lag.2"] <- glb_trnent_df[nrow(glb_trnent_df),
# "<col_name>"]
# glb_trnent_df <- mutate(glb_trnent_df,
# <new_col_name>=
# )
# glb_newent_df <- mutate(glb_newent_df,
# <new_col_name>=
# )
# print(summary(glb_trnent_df))
# print(summary(glb_newent_df))
# print(sapply(names(glb_trnent_df), function(col) sum(is.na(glb_trnent_df[, col]))))
# print(sapply(names(glb_newent_df), function(col) sum(is.na(glb_newent_df[, col]))))
# print(myplot_scatter(glb_trnent_df, "<col1_name>", "<col2_name>", smooth=TRUE))
replay.petrisim(pn=glb_analytics_pn,
replay.trans=(glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
"data.training.all","data.new")), flip_coord=TRUE)## time trans "bgn " "fit.data.training.all " "predict.data.new " "end "
## 0.0000 multiple enabled transitions: data.training.all data.new model.selected firing: data.training.all
## 1.0000 1 2 1 0 0
## 1.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction firing: data.new
## 2.0000 2 1 1 1 0
glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="select_features",
chunk_step_major=max(glb_script_df$chunk_step_major)+1,
chunk_step_minor=0,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed5 extract_features 3 0 6.361
## elapsed6 select_features 4 0 7.735Step 4: select features
print(glb_feats_df <- myselect_features(entity_df=glb_trnent_df,
exclude_vars_as_features=glb_exclude_vars_as_features,
rsp_var=glb_rsp_var))## id cor.y exclude.as.feat
## age age 0.24780278 0
## capitalgain capitalgain 0.23222327 0
## hoursperweek hoursperweek 0.22402106 0
## sex.fctr sex.fctr -0.20283667 0
## relationship.fctr relationship.fctr -0.12845800 0
## occupation.fctr occupation.fctr -0.12686813 0
## capitalloss capitalloss 0.10182189 0
## workclass.fctr workclass.fctr 0.07835390 0
## nativecountry.fctr nativecountry.fctr -0.06238053 0
## education.fctr education.fctr -0.06153968 0
## .rnorm .rnorm -0.01983209 0
## race.fctr race.fctr -0.01730292 0
## maritalstatus.fctr maritalstatus.fctr 0.01310876 0
## cor.y.abs
## age 0.24780278
## capitalgain 0.23222327
## hoursperweek 0.22402106
## sex.fctr 0.20283667
## relationship.fctr 0.12845800
## occupation.fctr 0.12686813
## capitalloss 0.10182189
## workclass.fctr 0.07835390
## nativecountry.fctr 0.06238053
## education.fctr 0.06153968
## .rnorm 0.01983209
## race.fctr 0.01730292
## maritalstatus.fctr 0.01310876glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="remove_correlated_features",
chunk_step_major=max(glb_script_df$chunk_step_major),
chunk_step_minor=glb_script_df[nrow(glb_script_df), "chunk_step_minor"]+1,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor
## elapsed6 select_features 4 0
## elapsed7 remove_correlated_features 4 1
## elapsed
## elapsed6 7.735
## elapsed7 7.936Step 5: fit models
if (glb_is_classification && glb_is_binomial && (length(unique(glb_trnent_df[, glb_rsp_var])) < 2))
stop("glb_trnent_df$", glb_rsp_var, ": contains less than 2 unique values: ", paste0(unique(glb_trnent_df[, glb_rsp_var]), collapse=", "))
max_cor_y_x_var <- orderBy(~ -cor.y.abs,
subset(glb_feats_df, (exclude.as.feat == 0) & (cor.low == 1)))[1, "id"]
if (!is.null(glb_Baseline_mdl_var)) {
if ((max_cor_y_x_var != glb_Baseline_mdl_var) &
(glb_feats_df[max_cor_y_x_var, "cor.y.abs"] >
glb_feats_df[glb_Baseline_mdl_var, "cor.y.abs"]))
stop(max_cor_y_x_var, " has a lower correlation with ", glb_rsp_var,
" than the Baseline var: ", glb_Baseline_mdl_var)
}
glb_model_type <- ifelse(glb_is_regression, "regression", "classification")
# Any models that have tuning parameters has "better" results with cross-validation (except rf)
# & "different" results for different outcome metrics
# Baseline
if (!is.null(glb_Baseline_mdl_var)) {
# lm_mdl <- lm(reformulate(glb_Baseline_mdl_var,
# response="bucket2009"), data=glb_trnent_df)
# print(summary(lm_mdl))
# plot(lm_mdl, ask=FALSE)
# ret_lst <- myfit_mdl_fn(model_id="Baseline",
# model_method=ifelse(glb_is_regression, "lm",
# ifelse(glb_is_binomial, "glm", "rpart")),
# indep_vars_vctr=glb_Baseline_mdl_var,
# rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
# fit_df=glb_trnent_df, OOB_df=glb_newent_df,
# n_cv_folds=0, tune_models_df=NULL,
# model_loss_mtrx=glb_model_metric_terms,
# model_summaryFunction=glb_model_metric_smmry,
# model_metric=glb_model_metric,
# model_metric_maximize=glb_model_metric_maximize)
ret_lst <- myfit_mdl_fn(model_id="Baseline", model_method="mybaseln_classfr",
indep_vars_vctr=glb_Baseline_mdl_var,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df)
}
# Most Frequent Outcome "MFO" model: mean(y) for regression
# Not using caret's nullModel since model stats not avl
# Cannot use rpart for multinomial classification since it predicts non-MFO
ret_lst <- myfit_mdl(model_id="MFO",
model_method=ifelse(glb_is_regression, "lm", "myMFO_classfr"),
model_type=glb_model_type,
indep_vars_vctr=".rnorm",
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df)## Loading required package: caret
## Loading required package: lattice
##
## Attaching package: 'caret'
##
## The following object is masked from 'package:survival':
##
## cluster## [1] "fitting model: MFO.myMFO_classfr"
## [1] " indep_vars: .rnorm"
## Fitting parameter = none on full training set
## [1] "in MFO.Classifier$fit"
## [1] "unique.vals:"
## [1] N Y
## Levels: N Y
## [1] "unique.prob:"
## y
## N Y
## 0.759 0.241
## [1] "MFO.val:"
## [1] "N"
## Length Class Mode
## unique.vals 2 factor numeric
## unique.prob 2 -none- numeric
## MFO.val 1 -none- character
## x.names 1 -none- character
## xNames 1 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character## Loading required package: ROCR
## Loading required package: gplots
##
## Attaching package: 'gplots'
##
## The following object is masked from 'package:stats':
##
## lowess## [1] "in MFO.Classifier$predict"
## [1] "in MFO.Classifier$prob"
## N Y
## 1 0.759 0.241
## 2 0.759 0.241
## 3 0.759 0.241
## 4 0.759 0.241
## 5 0.759 0.241
## 6 0.759 0.241
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.fit"
## over50k.fctr over50k.fctr.predict.MFO.myMFO_classfr.N
## 1 N 759
## 2 Y 241
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.MFO.myMFO_classfr.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.MFO.myMFO_classfr.Y
## 1 0
## 2 0
## Prediction
## Reference N Y
## N 759 0
## Y 241 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.590000e-01 0.000000e+00 7.312521e-01 7.852137e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 5.172861e-01 6.484063e-54
## [1] "in MFO.Classifier$predict"
## [1] "in MFO.Classifier$prob"
## N Y
## 1 0.759 0.241
## 2 0.759 0.241
## 3 0.759 0.241
## 4 0.759 0.241
## 5 0.759 0.241
## 6 0.759 0.241
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.OOB"
## over50k.fctr over50k.fctr.predict.MFO.myMFO_classfr.N
## 1 N 759
## 2 Y 241
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.MFO.myMFO_classfr.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.MFO.myMFO_classfr.Y
## 1 0
## 2 0
## Prediction
## Reference N Y
## N 759 0
## Y 241 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.590000e-01 0.000000e+00 7.312521e-01 7.852137e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 5.172861e-01 6.484063e-54
## model_id model_method feats max.nTuningRuns
## 1 MFO.myMFO_classfr myMFO_classfr .rnorm 0
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 0.372 0.002 0.5
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.5 0 0.759
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.7312521 0.7852137 0 0.5
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.5 0 0.759
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.7312521 0.7852137 0if (glb_is_classification)
# "random" model - only for classification; none needed for regression since it is same as MFO
ret_lst <- myfit_mdl(model_id="Random", model_method="myrandom_classfr",
model_type=glb_model_type,
indep_vars_vctr=".rnorm",
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df)## [1] "fitting model: Random.myrandom_classfr"
## [1] " indep_vars: .rnorm"
## Fitting parameter = none on full training set
## Length Class Mode
## unique.vals 2 factor numeric
## unique.prob 2 table numeric
## xNames 1 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character
## [1] "in Random.Classifier$prob"
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 578 183
## Y 181 58
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 578
## 2 Y 183
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 181
## 2 58
## Reference
## Prediction N Y
## N 578 183
## Y 181 58
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 578
## 2 Y 183
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 181
## 2 58
## Reference
## Prediction N Y
## N 578 183
## Y 181 58
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 578
## 2 Y 183
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 181
## 2 58
## Reference
## Prediction N Y
## N 578 183
## Y 181 58
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 578
## 2 Y 183
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 181
## 2 58
## Reference
## Prediction N Y
## N 578 183
## Y 181 58
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 578
## 2 Y 183
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 181
## 2 58
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.3883965
## 2 0.1 0.3883965
## 3 0.2 0.3883965
## 4 0.3 0.2416667
## 5 0.4 0.2416667
## 6 0.5 0.2416667
## 7 0.6 0.2416667
## 8 0.7 0.2416667
## 9 0.8 0.0000000
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.fit"
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 N 759
## 2 Y 241
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 759
## 2 241
## Prediction
## Reference N Y
## N 0 759
## Y 0 241
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 2.410000e-01 0.000000e+00 2.147863e-01 2.687479e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 1.205237e-166
## [1] "in Random.Classifier$prob"
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 558 183
## Y 201 58
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 558
## 2 Y 183
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 201
## 2 58
## Reference
## Prediction N Y
## N 558 183
## Y 201 58
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 558
## 2 Y 183
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 201
## 2 58
## Reference
## Prediction N Y
## N 558 183
## Y 201 58
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 558
## 2 Y 183
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 201
## 2 58
## Reference
## Prediction N Y
## N 558 183
## Y 201 58
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 558
## 2 Y 183
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 201
## 2 58
## Reference
## Prediction N Y
## N 558 183
## Y 201 58
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 558
## 2 Y 183
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 201
## 2 58
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.3883965
## 2 0.1 0.3883965
## 3 0.2 0.3883965
## 4 0.3 0.2320000
## 5 0.4 0.2320000
## 6 0.5 0.2320000
## 7 0.6 0.2320000
## 8 0.7 0.2320000
## 9 0.8 0.0000000
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.OOB"
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 N 759
## 2 Y 241
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.Random.myrandom_classfr.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.Random.myrandom_classfr.Y
## 1 759
## 2 241
## Prediction
## Reference N Y
## N 0 759
## Y 0 241
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 2.410000e-01 0.000000e+00 2.147863e-01 2.687479e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 1.205237e-166
## model_id model_method feats max.nTuningRuns
## 1 Random.myrandom_classfr myrandom_classfr .rnorm 0
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 0.223 0.001 0.5010961
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.2 0.3883965 0.241
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.2147863 0.2687479 0 0.4879209
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.2 0.3883965 0.241
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.2147863 0.2687479 0# Max.cor.Y
# Check impact of cv
# rpart is not a good candidate since caret does not optimize cp (only tuning parameter of rpart) well
ret_lst <- myfit_mdl(model_id="Max.cor.Y.cv.0",
model_method="rpart",
model_type=glb_model_type,
indep_vars_vctr=max_cor_y_x_var,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df)## [1] "fitting model: Max.cor.Y.cv.0.rpart"
## [1] " indep_vars: age"## Loading required package: rpart## Fitting cp = 0.00083 on full training set## Loading required package: rpart.plot
## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 1000
##
## CP nsplit rel error
## 1 0.0008298755 0 1
##
## Node number 1: 1000 observations
## predicted class=N expected loss=0.241 P(node) =1
## class counts: 759 241
## probabilities: 0.759 0.241
##
## n= 1000
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 1000 241 N (0.7590000 0.2410000) *
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.fit"
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.rpart.N
## 1 N 759
## 2 Y 241
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.rpart.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.cv.0.rpart.Y
## 1 0
## 2 0
## Prediction
## Reference N Y
## N 759 0
## Y 241 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.590000e-01 0.000000e+00 7.312521e-01 7.852137e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 5.172861e-01 6.484063e-54
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.OOB"
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.rpart.N
## 1 N 759
## 2 Y 241
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.rpart.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.cv.0.rpart.Y
## 1 0
## 2 0
## Prediction
## Reference N Y
## N 759 0
## Y 241 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.590000e-01 0.000000e+00 7.312521e-01 7.852137e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 5.172861e-01 6.484063e-54
## model_id model_method feats max.nTuningRuns
## 1 Max.cor.Y.cv.0.rpart rpart age 0
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 0.779 0.02 0.5
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.5 0 0.759
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.7312521 0.7852137 0 0.5
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.5 0 0.759
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.7312521 0.7852137 0ret_lst <- myfit_mdl(model_id="Max.cor.Y.cv.0.cp.0",
model_method="rpart",
model_type=glb_model_type,
indep_vars_vctr=max_cor_y_x_var,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df,
n_cv_folds=0, tune_models_df=data.frame(parameter="cp", min=0.0, max=0.0, by=0.1))## [1] "fitting model: Max.cor.Y.cv.0.cp.0.rpart"
## [1] " indep_vars: age"
## Fitting cp = 0 on full training set
## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 1000
##
## CP nsplit rel error
## 1 0.0008298755 0 1.0000000
## 2 0.0000000000 5 0.9958506
##
## Variable importance
## age
## 100
##
## Node number 1: 1000 observations, complexity param=0.0008298755
## predicted class=N expected loss=0.241 P(node) =1
## class counts: 759 241
## probabilities: 0.759 0.241
## left son=2 (256 obs) right son=3 (744 obs)
## Primary splits:
## age < 28.5 to the left, improve=33.75374, (0 missing)
##
## Node number 2: 256 observations
## predicted class=N expected loss=0.01953125 P(node) =0.256
## class counts: 251 5
## probabilities: 0.980 0.020
##
## Node number 3: 744 observations, complexity param=0.0008298755
## predicted class=N expected loss=0.3172043 P(node) =0.744
## class counts: 508 236
## probabilities: 0.683 0.317
## left son=6 (265 obs) right son=7 (479 obs)
## Primary splits:
## age < 38.5 to the left, improve=5.198796, (0 missing)
##
## Node number 6: 265 observations
## predicted class=N expected loss=0.2377358 P(node) =0.265
## class counts: 202 63
## probabilities: 0.762 0.238
##
## Node number 7: 479 observations, complexity param=0.0008298755
## predicted class=N expected loss=0.3611691 P(node) =0.479
## class counts: 306 173
## probabilities: 0.639 0.361
## left son=14 (129 obs) right son=15 (350 obs)
## Primary splits:
## age < 55.5 to the right, improve=2.379942, (0 missing)
##
## Node number 14: 129 observations
## predicted class=N expected loss=0.2790698 P(node) =0.129
## class counts: 93 36
## probabilities: 0.721 0.279
##
## Node number 15: 350 observations, complexity param=0.0008298755
## predicted class=N expected loss=0.3914286 P(node) =0.35
## class counts: 213 137
## probabilities: 0.609 0.391
## left son=30 (230 obs) right son=31 (120 obs)
## Primary splits:
## age < 47.5 to the left, improve=0.9217598, (0 missing)
##
## Node number 30: 230 observations
## predicted class=N expected loss=0.3652174 P(node) =0.23
## class counts: 146 84
## probabilities: 0.635 0.365
##
## Node number 31: 120 observations, complexity param=0.0008298755
## predicted class=N expected loss=0.4416667 P(node) =0.12
## class counts: 67 53
## probabilities: 0.558 0.442
## left son=62 (83 obs) right son=63 (37 obs)
## Primary splits:
## age < 49.5 to the right, improve=0.5522685, (0 missing)
##
## Node number 62: 83 observations
## predicted class=N expected loss=0.4096386 P(node) =0.083
## class counts: 49 34
## probabilities: 0.590 0.410
##
## Node number 63: 37 observations
## predicted class=Y expected loss=0.4864865 P(node) =0.037
## class counts: 18 19
## probabilities: 0.486 0.514
##
## n= 1000
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 1000 241 N (0.75900000 0.24100000)
## 2) age< 28.5 256 5 N (0.98046875 0.01953125) *
## 3) age>=28.5 744 236 N (0.68279570 0.31720430)
## 6) age< 38.5 265 63 N (0.76226415 0.23773585) *
## 7) age>=38.5 479 173 N (0.63883090 0.36116910)
## 14) age>=55.5 129 36 N (0.72093023 0.27906977) *
## 15) age< 55.5 350 137 N (0.60857143 0.39142857)
## 30) age< 47.5 230 84 N (0.63478261 0.36521739) *
## 31) age>=47.5 120 53 N (0.55833333 0.44166667)
## 62) age>=49.5 83 34 N (0.59036145 0.40963855) *
## 63) age< 49.5 37 18 Y (0.48648649 0.51351351) *
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 251 5
## Y 508 236
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 251
## 2 Y 5
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 508
## 2 236
## Reference
## Prediction N Y
## N 251 5
## Y 508 236
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 251
## 2 Y 5
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 508
## 2 236
## Reference
## Prediction N Y
## N 546 104
## Y 213 137
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 546
## 2 Y 104
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 213
## 2 137
## Reference
## Prediction N Y
## N 692 188
## Y 67 53
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 692
## 2 Y 188
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 67
## 2 53
## Reference
## Prediction N Y
## N 741 222
## Y 18 19
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 741
## 2 Y 222
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 18
## 2 19
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.3883965
## 2 0.1 0.4791878
## 3 0.2 0.4791878
## 4 0.3 0.4636210
## 5 0.4 0.2936288
## 6 0.5 0.1366906
## 7 0.6 0.0000000
## 8 0.7 0.0000000
## 9 0.8 0.0000000
## 10 0.9 0.0000000
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.fit"
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 251
## 2 Y 5
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 508
## 2 236
## Reference
## Prediction N Y
## N 251 5
## Y 508 236
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 251
## 2 Y 5
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 508
## 2 236
## Prediction
## Reference N Y
## N 251 508
## Y 5 236
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 4.870000e-01 1.810240e-01 4.555980e-01 5.184787e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 7.671886e-109
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 274 14
## Y 485 227
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 274
## 2 Y 14
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 485
## 2 227
## Reference
## Prediction N Y
## N 274 14
## Y 485 227
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 274
## 2 Y 14
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 485
## 2 227
## Reference
## Prediction N Y
## N 549 124
## Y 210 117
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 549
## 2 Y 124
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 210
## 2 117
## Reference
## Prediction N Y
## N 687 198
## Y 72 43
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 687
## 2 Y 198
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 72
## 2 43
## Reference
## Prediction N Y
## N 736 232
## Y 23 9
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 736
## 2 Y 232
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 23
## 2 9
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.38839645
## 2 0.1 0.47639035
## 3 0.2 0.47639035
## 4 0.3 0.41197183
## 5 0.4 0.24157303
## 6 0.5 0.06593407
## 7 0.6 0.00000000
## 8 0.7 0.00000000
## 9 0.8 0.00000000
## 10 0.9 0.00000000
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.OOB"
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 274
## 2 Y 14
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 485
## 2 227
## Reference
## Prediction N Y
## N 274 14
## Y 485 227
## over50k.fctr over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.N
## 1 N 274
## 2 Y 14
## over50k.fctr.predict.Max.cor.Y.cv.0.cp.0.rpart.Y
## 1 485
## 2 227
## Prediction
## Reference N Y
## N 274 485
## Y 14 227
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.010000e-01 1.817204e-01 4.695463e-01 5.324478e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 2.818448e-98
## model_id model_method feats max.nTuningRuns
## 1 Max.cor.Y.cv.0.cp.0.rpart rpart age 0
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 0.668 0.018 0.7183097
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.2 0.4791878 0.487
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.455598 0.5184787 0.181024 0.6837507
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.2 0.4763903 0.501
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.4695463 0.5324478 0.1817204if (glb_is_regression || glb_is_binomial) # For multinomials this model will be run next by default
ret_lst <- myfit_mdl(model_id="Max.cor.Y",
model_method="rpart",
model_type=glb_model_type,
indep_vars_vctr=max_cor_y_x_var,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)## [1] "fitting model: Max.cor.Y.rpart"
## [1] " indep_vars: age"
## + Fold1: cp=0
## - Fold1: cp=0
## + Fold2: cp=0
## - Fold2: cp=0
## + Fold3: cp=0
## - Fold3: cp=0
## Aggregating results
## Selecting tuning parameters
## Fitting cp = 0.00083 on full training set## Warning in myfit_mdl(model_id = "Max.cor.Y", model_method = "rpart",
## model_type = glb_model_type, : model's bestTune found at an extreme of
## tuneGrid for parameter: cp
## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 1000
##
## CP nsplit rel error
## 1 0.0008298755 0 1
##
## Node number 1: 1000 observations
## predicted class=N expected loss=0.241 P(node) =1
## class counts: 759 241
## probabilities: 0.759 0.241
##
## n= 1000
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 1000 241 N (0.7590000 0.2410000) *
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.fit"
## over50k.fctr over50k.fctr.predict.Max.cor.Y.rpart.N
## 1 N 759
## 2 Y 241
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.rpart.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.rpart.Y
## 1 0
## 2 0
## Prediction
## Reference N Y
## N 759 0
## Y 241 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.590000e-01 0.000000e+00 7.312521e-01 7.852137e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 5.172861e-01 6.484063e-54
## [1] "Classifier Probability Threshold: 0.5000 to maximize f.score.OOB"
## over50k.fctr over50k.fctr.predict.Max.cor.Y.rpart.N
## 1 N 759
## 2 Y 241
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.rpart.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.rpart.Y
## 1 0
## 2 0
## Prediction
## Reference N Y
## N 759 0
## Y 241 0
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.590000e-01 0.000000e+00 7.312521e-01 7.852137e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 5.172861e-01 6.484063e-54
## model_id model_method feats max.nTuningRuns
## 1 Max.cor.Y.rpart rpart age 3
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 0.978 0.021 0.5
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.5 0 0.7459975
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.7312521 0.7852137 0.052497 0.5
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.5 0 0.759
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.7312521 0.7852137 0
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.01220642 0.02765327# Used to compare vs. Interactions.High.cor.Y
ret_lst <- myfit_mdl(model_id="Max.cor.Y",
model_method=ifelse(glb_is_regression, "lm",
ifelse(glb_is_binomial, "glm", "rpart")),
model_type=glb_model_type,
indep_vars_vctr=max_cor_y_x_var,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)## [1] "fitting model: Max.cor.Y.glm"
## [1] " indep_vars: age"
## + Fold1: parameter=none
## - Fold1: parameter=none
## + Fold2: parameter=none
## - Fold2: parameter=none
## + Fold3: parameter=none
## - Fold3: parameter=none
## Aggregating results
## Fitting final model on full training set
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.5972 -0.7599 -0.5935 -0.4677 1.9655
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.881280 0.250694 -11.49 < 2e-16 ***
## age 0.042547 0.005628 7.56 4.02e-14 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1104.5 on 999 degrees of freedom
## Residual deviance: 1044.1 on 998 degrees of freedom
## AIC: 1048.1
##
## Number of Fisher Scoring iterations: 4
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 396 54
## Y 363 187
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 396
## 2 Y 54
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 363
## 2 187
## Reference
## Prediction N Y
## N 599 152
## Y 160 89
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 599
## 2 Y 152
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 160
## 2 89
## Reference
## Prediction N Y
## N 690 213
## Y 69 28
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 690
## 2 Y 213
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 69
## 2 28
## Reference
## Prediction N Y
## N 745 234
## Y 14 7
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 745
## 2 Y 234
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 14
## 2 7
## Reference
## Prediction N Y
## N 757 239
## Y 2 2
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 757
## 2 Y 239
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 2
## 2 2
## Reference
## Prediction N Y
## N 758 241
## Y 1 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 758
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 1
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.38839645
## 2 0.1 0.38839645
## 3 0.2 0.47281922
## 4 0.3 0.36326531
## 5 0.4 0.16568047
## 6 0.5 0.05343511
## 7 0.6 0.01632653
## 8 0.7 0.00000000
## 9 0.8 0.00000000
## 10 0.9 0.00000000
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.fit"
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 396
## 2 Y 54
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 363
## 2 187
## Reference
## Prediction N Y
## N 396 54
## Y 363 187
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 396
## 2 Y 54
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 363
## 2 187
## Prediction
## Reference N Y
## N 396 363
## Y 54 187
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 5.830000e-01 2.070736e-01 5.517330e-01 6.137769e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 2.100846e-51
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 411 49
## Y 348 192
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 411
## 2 Y 49
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 348
## 2 192
## Reference
## Prediction N Y
## N 597 153
## Y 162 88
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 597
## 2 Y 153
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 162
## 2 88
## Reference
## Prediction N Y
## N 694 207
## Y 65 34
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 694
## 2 Y 207
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 65
## 2 34
## Reference
## Prediction N Y
## N 737 236
## Y 22 5
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 737
## 2 Y 236
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 22
## 2 5
## Reference
## Prediction N Y
## N 753 241
## Y 6 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 753
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 6
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 0
## 2 0
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.38839645
## 2 0.1 0.38839645
## 3 0.2 0.49167734
## 4 0.3 0.35845214
## 5 0.4 0.20000000
## 6 0.5 0.03731343
## 7 0.6 0.00000000
## 8 0.7 0.00000000
## 9 0.8 0.00000000
## 10 0.9 0.00000000
## 11 1.0 0.00000000
## [1] "Classifier Probability Threshold: 0.2000 to maximize f.score.OOB"
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 411
## 2 Y 49
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 348
## 2 192
## Reference
## Prediction N Y
## N 411 49
## Y 348 192
## over50k.fctr over50k.fctr.predict.Max.cor.Y.glm.N
## 1 N 411
## 2 Y 49
## over50k.fctr.predict.Max.cor.Y.glm.Y
## 1 348
## 2 192
## Prediction
## Reference N Y
## N 411 348
## Y 49 192
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 6.030000e-01 2.375941e-01 5.719080e-01 6.334837e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 1.000000e+00 1.419192e-50
## model_id model_method feats max.nTuningRuns
## 1 Max.cor.Y.glm glm age 1
## min.elapsedtime.everything min.elapsedtime.final max.auc.fit
## 1 1.074 0.019 0.6880368
## opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.2 0.4728192 0.7550095
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.551733 0.6137769 0.02934095 0.6937196
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.2 0.4916773 0.603
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.571908 0.6334837 0.2375941 1048.081
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.008758591 0.03529993# Interactions.High.cor.Y
if (nrow(int_feats_df <- subset(glb_feats_df, (cor.low == 0) &
(exclude.as.feat == 0))) > 0) {
# lm & glm handle interaction terms; rpart & rf do not
# This does not work - why ???
# indep_vars_vctr <- ifelse(glb_is_binomial,
# c(max_cor_y_x_var, paste(max_cor_y_x_var,
# subset(glb_feats_df, is.na(cor.low))[, "id"], sep=":")),
# union(max_cor_y_x_var, subset(glb_feats_df, is.na(cor.low))[, "id"]))
if (glb_is_regression || glb_is_binomial) {
indep_vars_vctr <-
c(max_cor_y_x_var, paste(max_cor_y_x_var, int_feats_df[, "id"], sep=":"))
} else { indep_vars_vctr <- union(max_cor_y_x_var, int_feats_df[, "id"]) }
ret_lst <- myfit_mdl(model_id="Interact.High.cor.y",
model_method=ifelse(glb_is_regression, "lm",
ifelse(glb_is_binomial, "glm", "rpart")),
model_type=glb_model_type,
indep_vars_vctr,
glb_rsp_var, glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)
}
# Low.cor.X
ret_lst <- myfit_mdl(model_id="Low.cor.X",
model_method=ifelse(glb_is_regression, "lm",
ifelse(glb_is_binomial, "glm", "rpart")),
indep_vars_vctr=subset(glb_feats_df, cor.low == 1)[, "id"],
model_type=glb_model_type,
glb_rsp_var, glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL)## [1] "fitting model: Low.cor.X.glm"
## [1] " indep_vars: age, capitalgain, hoursperweek, capitalloss, workclass.fctr, maritalstatus.fctr, race.fctr, .rnorm, education.fctr, nativecountry.fctr, occupation.fctr, relationship.fctr, sex.fctr"
## + Fold1: parameter=none## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## - Fold1: parameter=none
## + Fold2: parameter=none## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## - Fold2: parameter=none
## + Fold3: parameter=none## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## - Fold3: parameter=none
## Aggregating results
## Fitting final model on full training set## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## Warning: not plotting observations with leverage one:
## 118, 160, 182, 194, 371, 422, 477, 518, 535, 638, 739, 798, 805, 808, 811, 921
## Warning: not plotting observations with leverage one:
## 118, 160, 182, 194, 371, 422, 477, 518, 535, 638, 739, 798, 805, 808, 811, 921
## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.24059 -0.43840 -0.08362 0.00000 2.62797
##
## Coefficients: (12 not defined because of singularities)
## Estimate Std. Error
## (Intercept) -6.821e+00 1.159e+00
## age 3.818e-02 1.089e-02
## capitalgain 4.161e-04 7.105e-05
## hoursperweek 2.970e-02 1.051e-02
## capitalloss 2.194e-04 2.357e-04
## `workclass.fctr Self-emp-not-inc` 1.610e-01 7.478e-01
## `workclass.fctr Private` 1.125e+00 6.789e-01
## `workclass.fctr Federal-gov` 2.904e+00 9.072e-01
## `workclass.fctr Local-gov` 3.428e-01 7.691e-01
## `workclass.fctr ?` 6.723e-01 9.573e-01
## `workclass.fctr Self-emp-inc` 1.042e+00 8.233e-01
## `workclass.fctr Without-pay` -1.725e+01 1.075e+04
## `workclass.fctr Never-worked` NA NA
## `maritalstatus.fctr Married-civ-spouse` -1.878e+00 7.345e+03
## `maritalstatus.fctr Divorced` 4.708e-01 5.824e-01
## `maritalstatus.fctr Married-spouse-absent` -1.558e+01 2.722e+03
## `maritalstatus.fctr Separated` 4.013e-01 8.565e-01
## `maritalstatus.fctr Married-AF-spouse` -1.751e+01 1.302e+04
## `maritalstatus.fctr Widowed` 1.088e+00 8.744e-01
## `race.fctr Black` 4.090e-01 4.460e-01
## `race.fctr Asian-Pac-Islander` 1.380e+00 1.491e+00
## `race.fctr Amer-Indian-Eskimo` 5.347e-01 1.206e+00
## `race.fctr Other` 1.267e+00 1.487e+00
## .rnorm -3.095e-02 1.171e-01
## `education.fctr HS-grad` -1.559e+00 3.436e-01
## `education.fctr 11th` -1.815e+01 2.222e+03
## `education.fctr Masters` -2.940e-02 4.506e-01
## `education.fctr 9th` -1.666e+00 1.406e+00
## `education.fctr Some-college` -8.444e-01 3.563e-01
## `education.fctr Assoc-acdm` -6.287e-01 7.288e-01
## `education.fctr 7th-8th` -2.846e+00 1.162e+00
## `education.fctr Doctorate` 1.534e+00 9.430e-01
## `education.fctr Assoc-voc` -8.237e-01 5.183e-01
## `education.fctr Prof-school` 1.781e+00 9.115e-01
## `education.fctr 5th-6th` -2.305e+00 1.545e+00
## `education.fctr 10th` -1.951e+00 7.429e-01
## `education.fctr 1st-4th` -2.408e+00 5.929e+03
## `education.fctr Preschool` -2.289e+00 1.097e+04
## `education.fctr 12th` -1.821e+01 2.665e+03
## `nativecountry.fctr Cuba` 2.854e+00 1.496e+00
## `nativecountry.fctr Jamaica` 1.769e-01 1.363e+00
## `nativecountry.fctr India` -1.501e+00 1.781e+00
## `nativecountry.fctr Mexico` -1.704e+01 2.148e+03
## `nativecountry.fctr South` 3.292e-01 1.099e+04
## `nativecountry.fctr Puerto-Rico` -1.631e+01 7.046e+03
## `nativecountry.fctr Honduras` NA NA
## `nativecountry.fctr England` 9.965e-01 1.034e+00
## `nativecountry.fctr Canada` 2.268e+00 2.148e+00
## `nativecountry.fctr Germany` -1.790e+01 4.090e+03
## `nativecountry.fctr Iran` NA NA
## `nativecountry.fctr Philippines` -1.854e+00 2.302e+00
## `nativecountry.fctr Italy` -1.582e+01 1.075e+04
## `nativecountry.fctr Poland` -1.698e+01 1.075e+04
## `nativecountry.fctr Columbia` -1.694e+01 4.154e+03
## `nativecountry.fctr Cambodia` NA NA
## `nativecountry.fctr Thailand` NA NA
## `nativecountry.fctr Ecuador` -1.120e+01 1.075e+04
## `nativecountry.fctr Laos` 5.683e-01 2.207e+00
## `nativecountry.fctr Taiwan` 1.847e+01 7.390e+03
## `nativecountry.fctr Haiti` -1.274e+01 1.075e+04
## `nativecountry.fctr Portugal` -1.658e+01 1.075e+04
## `nativecountry.fctr Dominican-Republic` NA NA
## `nativecountry.fctr El-Salvador` -1.699e+01 3.705e+03
## `nativecountry.fctr France` -2.093e+01 6.381e+03
## `nativecountry.fctr Guatemala` 1.553e+01 1.175e+04
## `nativecountry.fctr China` -2.950e+00 2.128e+00
## `nativecountry.fctr Japan` -1.679e+01 1.075e+04
## `nativecountry.fctr Yugoslavia` NA NA
## `nativecountry.fctr Peru` -1.283e+01 1.075e+04
## `nativecountry.fctr Outlying-US(Guam-USVI-etc)` NA NA
## `nativecountry.fctr Scotland` -1.651e+01 1.075e+04
## `nativecountry.fctr Trinadad&Tobago` -1.849e+01 1.075e+04
## `nativecountry.fctr Greece` -1.473e+01 1.075e+04
## `nativecountry.fctr Nicaragua` -1.745e+01 1.075e+04
## `nativecountry.fctr Vietnam` -1.078e+00 7.665e+03
## `nativecountry.fctr Hong` -1.907e+01 1.075e+04
## `nativecountry.fctr Ireland` -1.925e+01 1.075e+04
## `nativecountry.fctr Hungary` NA NA
## `nativecountry.fctr Holand-Netherlands` NA NA
## `occupation.fctr Exec-managerial` 1.304e+00 4.950e-01
## `occupation.fctr Handlers-cleaners` 2.694e-02 8.208e-01
## `occupation.fctr Prof-specialty` 1.153e+00 5.194e-01
## `occupation.fctr Other-service` -2.126e+00 1.190e+00
## `occupation.fctr Sales` 1.069e+00 5.227e-01
## `occupation.fctr Transport-moving` 1.505e-01 6.237e-01
## `occupation.fctr Farming-fishing` -1.718e+01 1.838e+03
## `occupation.fctr Machine-op-inspct` 3.740e-01 6.315e-01
## `occupation.fctr Tech-support` 1.914e+00 7.198e-01
## `occupation.fctr Craft-repair` 4.127e-01 5.182e-01
## `occupation.fctr ?` NA NA
## `occupation.fctr Protective-serv` -1.078e+00 1.275e+00
## `occupation.fctr Armed-Forces` NA NA
## `occupation.fctr Priv-house-serv` -1.410e+01 4.167e+03
## `relationship.fctr Husband` 4.541e+00 7.345e+03
## `relationship.fctr Wife` 6.133e+00 7.345e+03
## `relationship.fctr Own-child` -9.141e-01 1.018e+00
## `relationship.fctr Unmarried` 1.153e+00 5.278e-01
## `relationship.fctr Other-relative` -1.367e+01 2.263e+03
## `sex.fctr Female` -1.083e+00 4.967e-01
## z value Pr(>|z|)
## (Intercept) -5.886 3.95e-09 ***
## age 3.507 0.000454 ***
## capitalgain 5.857 4.71e-09 ***
## hoursperweek 2.828 0.004691 **
## capitalloss 0.931 0.351975
## `workclass.fctr Self-emp-not-inc` 0.215 0.829567
## `workclass.fctr Private` 1.657 0.097516 .
## `workclass.fctr Federal-gov` 3.201 0.001370 **
## `workclass.fctr Local-gov` 0.446 0.655834
## `workclass.fctr ?` 0.702 0.482461
## `workclass.fctr Self-emp-inc` 1.265 0.205836
## `workclass.fctr Without-pay` -0.002 0.998720
## `workclass.fctr Never-worked` NA NA
## `maritalstatus.fctr Married-civ-spouse` 0.000 0.999796
## `maritalstatus.fctr Divorced` 0.808 0.418880
## `maritalstatus.fctr Married-spouse-absent` -0.006 0.995434
## `maritalstatus.fctr Separated` 0.469 0.639391
## `maritalstatus.fctr Married-AF-spouse` -0.001 0.998927
## `maritalstatus.fctr Widowed` 1.244 0.213532
## `race.fctr Black` 0.917 0.359078
## `race.fctr Asian-Pac-Islander` 0.925 0.354892
## `race.fctr Amer-Indian-Eskimo` 0.443 0.657450
## `race.fctr Other` 0.852 0.394420
## .rnorm -0.264 0.791562
## `education.fctr HS-grad` -4.537 5.71e-06 ***
## `education.fctr 11th` -0.008 0.993482
## `education.fctr Masters` -0.065 0.947977
## `education.fctr 9th` -1.185 0.236034
## `education.fctr Some-college` -2.370 0.017790 *
## `education.fctr Assoc-acdm` -0.863 0.388343
## `education.fctr 7th-8th` -2.450 0.014284 *
## `education.fctr Doctorate` 1.627 0.103707
## `education.fctr Assoc-voc` -1.589 0.112020
## `education.fctr Prof-school` 1.954 0.050691 .
## `education.fctr 5th-6th` -1.492 0.135688
## `education.fctr 10th` -2.626 0.008635 **
## `education.fctr 1st-4th` 0.000 0.999676
## `education.fctr Preschool` 0.000 0.999833
## `education.fctr 12th` -0.007 0.994550
## `nativecountry.fctr Cuba` 1.908 0.056379 .
## `nativecountry.fctr Jamaica` 0.130 0.896757
## `nativecountry.fctr India` -0.843 0.399391
## `nativecountry.fctr Mexico` -0.008 0.993670
## `nativecountry.fctr South` 0.000 0.999976
## `nativecountry.fctr Puerto-Rico` -0.002 0.998153
## `nativecountry.fctr Honduras` NA NA
## `nativecountry.fctr England` 0.964 0.334943
## `nativecountry.fctr Canada` 1.056 0.291095
## `nativecountry.fctr Germany` -0.004 0.996508
## `nativecountry.fctr Iran` NA NA
## `nativecountry.fctr Philippines` -0.805 0.420791
## `nativecountry.fctr Italy` -0.001 0.998826
## `nativecountry.fctr Poland` -0.002 0.998740
## `nativecountry.fctr Columbia` -0.004 0.996746
## `nativecountry.fctr Cambodia` NA NA
## `nativecountry.fctr Thailand` NA NA
## `nativecountry.fctr Ecuador` -0.001 0.999169
## `nativecountry.fctr Laos` 0.258 0.796765
## `nativecountry.fctr Taiwan` 0.002 0.998006
## `nativecountry.fctr Haiti` -0.001 0.999054
## `nativecountry.fctr Portugal` -0.002 0.998770
## `nativecountry.fctr Dominican-Republic` NA NA
## `nativecountry.fctr El-Salvador` -0.005 0.996341
## `nativecountry.fctr France` -0.003 0.997383
## `nativecountry.fctr Guatemala` 0.001 0.998945
## `nativecountry.fctr China` -1.386 0.165625
## `nativecountry.fctr Japan` -0.002 0.998754
## `nativecountry.fctr Yugoslavia` NA NA
## `nativecountry.fctr Peru` -0.001 0.999048
## `nativecountry.fctr Outlying-US(Guam-USVI-etc)` NA NA
## `nativecountry.fctr Scotland` -0.002 0.998775
## `nativecountry.fctr Trinadad&Tobago` -0.002 0.998628
## `nativecountry.fctr Greece` -0.001 0.998907
## `nativecountry.fctr Nicaragua` -0.002 0.998705
## `nativecountry.fctr Vietnam` 0.000 0.999888
## `nativecountry.fctr Hong` -0.002 0.998585
## `nativecountry.fctr Ireland` -0.002 0.998572
## `nativecountry.fctr Hungary` NA NA
## `nativecountry.fctr Holand-Netherlands` NA NA
## `occupation.fctr Exec-managerial` 2.634 0.008448 **
## `occupation.fctr Handlers-cleaners` 0.033 0.973816
## `occupation.fctr Prof-specialty` 2.220 0.026412 *
## `occupation.fctr Other-service` -1.786 0.074026 .
## `occupation.fctr Sales` 2.045 0.040865 *
## `occupation.fctr Transport-moving` 0.241 0.809342
## `occupation.fctr Farming-fishing` -0.009 0.992542
## `occupation.fctr Machine-op-inspct` 0.592 0.553673
## `occupation.fctr Tech-support` 2.658 0.007851 **
## `occupation.fctr Craft-repair` 0.796 0.425807
## `occupation.fctr ?` NA NA
## `occupation.fctr Protective-serv` -0.846 0.397550
## `occupation.fctr Armed-Forces` NA NA
## `occupation.fctr Priv-house-serv` -0.003 0.997300
## `relationship.fctr Husband` 0.001 0.999507
## `relationship.fctr Wife` 0.001 0.999334
## `relationship.fctr Own-child` -0.898 0.369177
## `relationship.fctr Unmarried` 2.183 0.029005 *
## `relationship.fctr Other-relative` -0.006 0.995180
## `sex.fctr Female` -2.180 0.029227 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1104.46 on 999 degrees of freedom
## Residual deviance: 545.16 on 913 degrees of freedom
## AIC: 719.16
##
## Number of Fisher Scoring iterations: 18## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 523 9
## Y 236 232
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 523
## 2 Y 9
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 236
## 2 232
## Reference
## Prediction N Y
## N 599 25
## Y 160 216
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 599
## 2 Y 25
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 160
## 2 216
## Reference
## Prediction N Y
## N 656 41
## Y 103 200
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 656
## 2 Y 41
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 103
## 2 200
## Reference
## Prediction N Y
## N 684 57
## Y 75 184
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 684
## 2 Y 57
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 75
## 2 184
## Reference
## Prediction N Y
## N 701 82
## Y 58 159
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 701
## 2 Y 82
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 58
## 2 159
## Reference
## Prediction N Y
## N 728 96
## Y 31 145
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 728
## 2 Y 96
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 31
## 2 145
## Reference
## Prediction N Y
## N 741 123
## Y 18 118
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 741
## 2 Y 123
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 18
## 2 118
## Reference
## Prediction N Y
## N 753 158
## Y 6 83
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 753
## 2 Y 158
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 6
## 2 83
## Reference
## Prediction N Y
## N 758 183
## Y 1 58
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 758
## 2 Y 183
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 1
## 2 58
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.3883965
## 2 0.1 0.6544429
## 3 0.2 0.7001621
## 4 0.3 0.7352941
## 5 0.4 0.7360000
## 6 0.5 0.6943231
## 7 0.6 0.6954436
## 8 0.7 0.6259947
## 9 0.8 0.5030303
## 10 0.9 0.3866667
## 11 1.0 0.0000000## [1] "Classifier Probability Threshold: 0.4000 to maximize f.score.fit"
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 684
## 2 Y 57
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 75
## 2 184
## Reference
## Prediction N Y
## N 684 57
## Y 75 184
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 684
## 2 Y 57
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 75
## 2 184
## Prediction
## Reference N Y
## N 684 75
## Y 57 184
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.680000e-01 6.481520e-01 8.454346e-01 8.883720e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 5.968845e-18 1.389640e-01## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 519 36
## Y 240 205
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 519
## 2 Y 36
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 240
## 2 205
## Reference
## Prediction N Y
## N 601 56
## Y 158 185
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 601
## 2 Y 56
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 158
## 2 185
## Reference
## Prediction N Y
## N 641 75
## Y 118 166
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 641
## 2 Y 75
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 118
## 2 166
## Reference
## Prediction N Y
## N 671 86
## Y 88 155
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 671
## 2 Y 86
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 88
## 2 155
## Reference
## Prediction N Y
## N 694 99
## Y 65 142
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 694
## 2 Y 99
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 65
## 2 142
## Reference
## Prediction N Y
## N 715 116
## Y 44 125
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 715
## 2 Y 116
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 44
## 2 125
## Reference
## Prediction N Y
## N 730 139
## Y 29 102
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 730
## 2 Y 139
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 29
## 2 102
## Reference
## Prediction N Y
## N 740 167
## Y 19 74
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 740
## 2 Y 167
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 19
## 2 74
## Reference
## Prediction N Y
## N 747 192
## Y 12 49
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 747
## 2 Y 192
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 12
## 2 49
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.3883965
## 2 0.1 0.5976676
## 3 0.2 0.6335616
## 4 0.3 0.6323810
## 5 0.4 0.6404959
## 6 0.5 0.6339286
## 7 0.6 0.6097561
## 8 0.7 0.5483871
## 9 0.8 0.4431138
## 10 0.9 0.3245033
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.4000 to maximize f.score.OOB"
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 671
## 2 Y 86
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 88
## 2 155
## Reference
## Prediction N Y
## N 671 86
## Y 88 155
## over50k.fctr over50k.fctr.predict.Low.cor.X.glm.N
## 1 N 671
## 2 Y 86
## over50k.fctr.predict.Low.cor.X.glm.Y
## 1 88
## 2 155
## Prediction
## Reference N Y
## N 671 88
## Y 86 155
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.260000e-01 5.257227e-01 8.010531e-01 8.490099e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 1.789552e-07 9.395704e-01
## model_id model_method
## 1 Low.cor.X.glm glm
## feats
## 1 age, capitalgain, hoursperweek, capitalloss, workclass.fctr, maritalstatus.fctr, race.fctr, .rnorm, education.fctr, nativecountry.fctr, occupation.fctr, relationship.fctr, sex.fctr
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 2.83 0.499
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.9325658 0.4 0.736 0.8299947
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.8454346 0.888372 0.5291816 0.835337
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.4 0.6404959 0.826
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.8010531 0.8490099 0.5257227 719.1617
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.05275524 0.1292318# User specified
for (method in glb_models_method_vctr) {
print(sprintf("iterating over method:%s", method))
# All X that is not user excluded
indep_vars_vctr <- setdiff(names(glb_trnent_df),
union(glb_rsp_var, glb_exclude_vars_as_features))
# easier to exclude features
# indep_vars_vctr <- setdiff(names(glb_trnent_df),
# union(union(glb_rsp_var, glb_exclude_vars_as_features),
# c("<feat1_name>", "<feat2_name>")))
# easier to include features
# indep_vars_vctr <- c("<feat1_name>", "<feat2_name>")
# User specified bivariate models
# indep_vars_vctr_lst <- list()
# for (feat in setdiff(names(glb_trnent_df),
# union(glb_rsp_var, glb_exclude_vars_as_features)))
# indep_vars_vctr_lst[["feat"]] <- feat
# User specified combinatorial models
# indep_vars_vctr_lst <- list()
# combn_mtrx <- combn(c("<feat1_name>", "<feat2_name>", "<featn_name>"),
# <num_feats_to_choose>)
# for (combn_ix in 1:ncol(combn_mtrx))
# #print(combn_mtrx[, combn_ix])
# indep_vars_vctr_lst[[combn_ix]] <- combn_mtrx[, combn_ix]
# glb_sel_mdl <- glb_sel_wlm_mdl <- ret_lst[["model"]]
# rpart_sel_wlm_mdl <- rpart(reformulate(indep_vars_vctr, response=glb_rsp_var),
# data=glb_trnent_df, method="class",
# control=rpart.control(cp=glb_sel_wlm_mdl$bestTune$cp),
# parms=list(loss=glb_model_metric_terms))
# print("rpart_sel_wlm_mdl"); prp(rpart_sel_wlm_mdl)
#
model_id_pfx <- "All.X";
ret_lst <- myfit_mdl(model_id=paste0(model_id_pfx, ""), model_method=method,
indep_vars_vctr=indep_vars_vctr,
model_type=glb_model_type,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=glb_tune_models_df)
# Since caret does not optimize rpart well
if (method == "rpart")
ret_lst <- myfit_mdl(model_id=paste0(model_id_pfx, ".cp.0"), model_method=method,
indep_vars_vctr=indep_vars_vctr,
model_type=glb_model_type,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df,
n_cv_folds=0, tune_models_df=data.frame(parameter="cp", min=0.0, max=0.0, by=0.1))
# Compare how rf performs w/i & w/o .rnorm
if (method == "rf")
ret_lst <- myfit_mdl(model_id=paste0(model_id_pfx, ".no.rnorm"), model_method=method,
indep_vars_vctr=setdiff(indep_vars_vctr, c(".rnorm")),
model_type=glb_model_type,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=glb_newent_df,
n_cv_folds=glb_n_cv_folds, tune_models_df=glb_tune_models_df)
# rf is hard-coded in caret to recognize only Accuracy / Kappa evaluation metrics
# only for OOB in trainControl ?
# ret_lst <- myfit_mdl_fn(model_id=paste0(model_id_pfx, ""), model_method=method,
# indep_vars_vctr=indep_vars_vctr,
# rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
# fit_df=glb_trnent_df, OOB_df=glb_newent_df,
# n_cv_folds=glb_n_cv_folds, tune_models_df=glb_tune_models_df,
# model_loss_mtrx=glb_model_metric_terms,
# model_summaryFunction=glb_model_metric_smmry,
# model_metric=glb_model_metric,
# model_metric_maximize=glb_model_metric_maximize)
}## [1] "iterating over method:glm"
## [1] "fitting model: All.X.glm"
## [1] " indep_vars: age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr, .rnorm"
## + Fold1: parameter=none## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## - Fold1: parameter=none
## + Fold2: parameter=none## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## - Fold2: parameter=none
## + Fold3: parameter=none## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## - Fold3: parameter=none
## Aggregating results
## Fitting final model on full training set## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## Warning: not plotting observations with leverage one:
## 118, 160, 182, 194, 239, 293, 371, 422, 477, 518, 535, 638, 739, 798, 805, 808, 811, 921 
## Warning: not plotting observations with leverage one:
## 118, 160, 182, 194, 239, 293, 371, 422, 477, 518, 535, 638, 739, 798, 805, 808, 811, 921
## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.24059 -0.43840 -0.08362 0.00000 2.62797
##
## Coefficients: (12 not defined because of singularities)
## Estimate Std. Error
## (Intercept) -6.821e+00 1.159e+00
## age 3.818e-02 1.089e-02
## capitalgain 4.161e-04 7.105e-05
## capitalloss 2.194e-04 2.357e-04
## hoursperweek 2.970e-02 1.051e-02
## `workclass.fctr Self-emp-not-inc` 1.610e-01 7.478e-01
## `workclass.fctr Private` 1.125e+00 6.789e-01
## `workclass.fctr Federal-gov` 2.904e+00 9.072e-01
## `workclass.fctr Local-gov` 3.428e-01 7.691e-01
## `workclass.fctr ?` 6.723e-01 9.573e-01
## `workclass.fctr Self-emp-inc` 1.042e+00 8.233e-01
## `workclass.fctr Without-pay` -1.725e+01 1.075e+04
## `workclass.fctr Never-worked` NA NA
## `education.fctr HS-grad` -1.559e+00 3.436e-01
## `education.fctr 11th` -1.815e+01 2.222e+03
## `education.fctr Masters` -2.940e-02 4.506e-01
## `education.fctr 9th` -1.666e+00 1.406e+00
## `education.fctr Some-college` -8.444e-01 3.563e-01
## `education.fctr Assoc-acdm` -6.287e-01 7.288e-01
## `education.fctr 7th-8th` -2.846e+00 1.162e+00
## `education.fctr Doctorate` 1.534e+00 9.430e-01
## `education.fctr Assoc-voc` -8.237e-01 5.183e-01
## `education.fctr Prof-school` 1.781e+00 9.115e-01
## `education.fctr 5th-6th` -2.305e+00 1.545e+00
## `education.fctr 10th` -1.951e+00 7.429e-01
## `education.fctr 1st-4th` -2.408e+00 5.929e+03
## `education.fctr Preschool` -2.289e+00 1.097e+04
## `education.fctr 12th` -1.821e+01 2.665e+03
## `maritalstatus.fctr Married-civ-spouse` -1.878e+00 7.345e+03
## `maritalstatus.fctr Divorced` 4.708e-01 5.824e-01
## `maritalstatus.fctr Married-spouse-absent` -1.558e+01 2.722e+03
## `maritalstatus.fctr Separated` 4.013e-01 8.565e-01
## `maritalstatus.fctr Married-AF-spouse` -1.751e+01 1.302e+04
## `maritalstatus.fctr Widowed` 1.088e+00 8.744e-01
## `occupation.fctr Exec-managerial` 1.304e+00 4.950e-01
## `occupation.fctr Handlers-cleaners` 2.694e-02 8.208e-01
## `occupation.fctr Prof-specialty` 1.153e+00 5.194e-01
## `occupation.fctr Other-service` -2.126e+00 1.190e+00
## `occupation.fctr Sales` 1.069e+00 5.227e-01
## `occupation.fctr Transport-moving` 1.505e-01 6.237e-01
## `occupation.fctr Farming-fishing` -1.718e+01 1.838e+03
## `occupation.fctr Machine-op-inspct` 3.740e-01 6.315e-01
## `occupation.fctr Tech-support` 1.914e+00 7.198e-01
## `occupation.fctr Craft-repair` 4.127e-01 5.182e-01
## `occupation.fctr ?` NA NA
## `occupation.fctr Protective-serv` -1.078e+00 1.275e+00
## `occupation.fctr Armed-Forces` NA NA
## `occupation.fctr Priv-house-serv` -1.410e+01 4.167e+03
## `relationship.fctr Husband` 4.541e+00 7.345e+03
## `relationship.fctr Wife` 6.133e+00 7.345e+03
## `relationship.fctr Own-child` -9.141e-01 1.018e+00
## `relationship.fctr Unmarried` 1.153e+00 5.278e-01
## `relationship.fctr Other-relative` -1.367e+01 2.263e+03
## `race.fctr Black` 4.090e-01 4.460e-01
## `race.fctr Asian-Pac-Islander` 1.380e+00 1.491e+00
## `race.fctr Amer-Indian-Eskimo` 5.347e-01 1.206e+00
## `race.fctr Other` 1.267e+00 1.487e+00
## `sex.fctr Female` -1.083e+00 4.967e-01
## `nativecountry.fctr Cuba` 2.854e+00 1.496e+00
## `nativecountry.fctr Jamaica` 1.769e-01 1.363e+00
## `nativecountry.fctr India` -1.501e+00 1.781e+00
## `nativecountry.fctr Mexico` -1.704e+01 2.148e+03
## `nativecountry.fctr South` 3.292e-01 1.099e+04
## `nativecountry.fctr Puerto-Rico` -1.631e+01 7.046e+03
## `nativecountry.fctr Honduras` NA NA
## `nativecountry.fctr England` 9.965e-01 1.034e+00
## `nativecountry.fctr Canada` 2.268e+00 2.148e+00
## `nativecountry.fctr Germany` -1.790e+01 4.090e+03
## `nativecountry.fctr Iran` NA NA
## `nativecountry.fctr Philippines` -1.854e+00 2.302e+00
## `nativecountry.fctr Italy` -1.582e+01 1.075e+04
## `nativecountry.fctr Poland` -1.698e+01 1.075e+04
## `nativecountry.fctr Columbia` -1.694e+01 4.154e+03
## `nativecountry.fctr Cambodia` NA NA
## `nativecountry.fctr Thailand` NA NA
## `nativecountry.fctr Ecuador` -1.120e+01 1.075e+04
## `nativecountry.fctr Laos` 5.683e-01 2.207e+00
## `nativecountry.fctr Taiwan` 1.847e+01 7.390e+03
## `nativecountry.fctr Haiti` -1.274e+01 1.075e+04
## `nativecountry.fctr Portugal` -1.658e+01 1.075e+04
## `nativecountry.fctr Dominican-Republic` NA NA
## `nativecountry.fctr El-Salvador` -1.699e+01 3.705e+03
## `nativecountry.fctr France` -2.093e+01 6.381e+03
## `nativecountry.fctr Guatemala` 1.553e+01 1.175e+04
## `nativecountry.fctr China` -2.950e+00 2.128e+00
## `nativecountry.fctr Japan` -1.679e+01 1.075e+04
## `nativecountry.fctr Yugoslavia` NA NA
## `nativecountry.fctr Peru` -1.283e+01 1.075e+04
## `nativecountry.fctr Outlying-US(Guam-USVI-etc)` NA NA
## `nativecountry.fctr Scotland` -1.651e+01 1.075e+04
## `nativecountry.fctr Trinadad&Tobago` -1.849e+01 1.075e+04
## `nativecountry.fctr Greece` -1.473e+01 1.075e+04
## `nativecountry.fctr Nicaragua` -1.745e+01 1.075e+04
## `nativecountry.fctr Vietnam` -1.078e+00 7.665e+03
## `nativecountry.fctr Hong` -1.907e+01 1.075e+04
## `nativecountry.fctr Ireland` -1.925e+01 1.075e+04
## `nativecountry.fctr Hungary` NA NA
## `nativecountry.fctr Holand-Netherlands` NA NA
## .rnorm -3.095e-02 1.171e-01
## z value Pr(>|z|)
## (Intercept) -5.886 3.95e-09 ***
## age 3.507 0.000454 ***
## capitalgain 5.857 4.71e-09 ***
## capitalloss 0.931 0.351975
## hoursperweek 2.828 0.004691 **
## `workclass.fctr Self-emp-not-inc` 0.215 0.829567
## `workclass.fctr Private` 1.657 0.097516 .
## `workclass.fctr Federal-gov` 3.201 0.001370 **
## `workclass.fctr Local-gov` 0.446 0.655834
## `workclass.fctr ?` 0.702 0.482461
## `workclass.fctr Self-emp-inc` 1.265 0.205836
## `workclass.fctr Without-pay` -0.002 0.998720
## `workclass.fctr Never-worked` NA NA
## `education.fctr HS-grad` -4.537 5.71e-06 ***
## `education.fctr 11th` -0.008 0.993482
## `education.fctr Masters` -0.065 0.947977
## `education.fctr 9th` -1.185 0.236034
## `education.fctr Some-college` -2.370 0.017790 *
## `education.fctr Assoc-acdm` -0.863 0.388343
## `education.fctr 7th-8th` -2.450 0.014284 *
## `education.fctr Doctorate` 1.627 0.103707
## `education.fctr Assoc-voc` -1.589 0.112020
## `education.fctr Prof-school` 1.954 0.050691 .
## `education.fctr 5th-6th` -1.492 0.135688
## `education.fctr 10th` -2.626 0.008635 **
## `education.fctr 1st-4th` 0.000 0.999676
## `education.fctr Preschool` 0.000 0.999833
## `education.fctr 12th` -0.007 0.994550
## `maritalstatus.fctr Married-civ-spouse` 0.000 0.999796
## `maritalstatus.fctr Divorced` 0.808 0.418880
## `maritalstatus.fctr Married-spouse-absent` -0.006 0.995434
## `maritalstatus.fctr Separated` 0.469 0.639391
## `maritalstatus.fctr Married-AF-spouse` -0.001 0.998927
## `maritalstatus.fctr Widowed` 1.244 0.213532
## `occupation.fctr Exec-managerial` 2.634 0.008448 **
## `occupation.fctr Handlers-cleaners` 0.033 0.973816
## `occupation.fctr Prof-specialty` 2.220 0.026412 *
## `occupation.fctr Other-service` -1.786 0.074026 .
## `occupation.fctr Sales` 2.045 0.040865 *
## `occupation.fctr Transport-moving` 0.241 0.809342
## `occupation.fctr Farming-fishing` -0.009 0.992542
## `occupation.fctr Machine-op-inspct` 0.592 0.553673
## `occupation.fctr Tech-support` 2.658 0.007851 **
## `occupation.fctr Craft-repair` 0.796 0.425807
## `occupation.fctr ?` NA NA
## `occupation.fctr Protective-serv` -0.846 0.397550
## `occupation.fctr Armed-Forces` NA NA
## `occupation.fctr Priv-house-serv` -0.003 0.997300
## `relationship.fctr Husband` 0.001 0.999507
## `relationship.fctr Wife` 0.001 0.999334
## `relationship.fctr Own-child` -0.898 0.369177
## `relationship.fctr Unmarried` 2.183 0.029005 *
## `relationship.fctr Other-relative` -0.006 0.995180
## `race.fctr Black` 0.917 0.359078
## `race.fctr Asian-Pac-Islander` 0.925 0.354892
## `race.fctr Amer-Indian-Eskimo` 0.443 0.657450
## `race.fctr Other` 0.852 0.394420
## `sex.fctr Female` -2.180 0.029227 *
## `nativecountry.fctr Cuba` 1.908 0.056379 .
## `nativecountry.fctr Jamaica` 0.130 0.896757
## `nativecountry.fctr India` -0.843 0.399391
## `nativecountry.fctr Mexico` -0.008 0.993670
## `nativecountry.fctr South` 0.000 0.999976
## `nativecountry.fctr Puerto-Rico` -0.002 0.998153
## `nativecountry.fctr Honduras` NA NA
## `nativecountry.fctr England` 0.964 0.334943
## `nativecountry.fctr Canada` 1.056 0.291095
## `nativecountry.fctr Germany` -0.004 0.996508
## `nativecountry.fctr Iran` NA NA
## `nativecountry.fctr Philippines` -0.805 0.420791
## `nativecountry.fctr Italy` -0.001 0.998826
## `nativecountry.fctr Poland` -0.002 0.998740
## `nativecountry.fctr Columbia` -0.004 0.996746
## `nativecountry.fctr Cambodia` NA NA
## `nativecountry.fctr Thailand` NA NA
## `nativecountry.fctr Ecuador` -0.001 0.999169
## `nativecountry.fctr Laos` 0.258 0.796765
## `nativecountry.fctr Taiwan` 0.002 0.998006
## `nativecountry.fctr Haiti` -0.001 0.999054
## `nativecountry.fctr Portugal` -0.002 0.998770
## `nativecountry.fctr Dominican-Republic` NA NA
## `nativecountry.fctr El-Salvador` -0.005 0.996341
## `nativecountry.fctr France` -0.003 0.997383
## `nativecountry.fctr Guatemala` 0.001 0.998945
## `nativecountry.fctr China` -1.386 0.165625
## `nativecountry.fctr Japan` -0.002 0.998754
## `nativecountry.fctr Yugoslavia` NA NA
## `nativecountry.fctr Peru` -0.001 0.999048
## `nativecountry.fctr Outlying-US(Guam-USVI-etc)` NA NA
## `nativecountry.fctr Scotland` -0.002 0.998775
## `nativecountry.fctr Trinadad&Tobago` -0.002 0.998628
## `nativecountry.fctr Greece` -0.001 0.998907
## `nativecountry.fctr Nicaragua` -0.002 0.998705
## `nativecountry.fctr Vietnam` 0.000 0.999888
## `nativecountry.fctr Hong` -0.002 0.998585
## `nativecountry.fctr Ireland` -0.002 0.998572
## `nativecountry.fctr Hungary` NA NA
## `nativecountry.fctr Holand-Netherlands` NA NA
## .rnorm -0.264 0.791562
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1104.46 on 999 degrees of freedom
## Residual deviance: 545.16 on 913 degrees of freedom
## AIC: 719.16
##
## Number of Fisher Scoring iterations: 18## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.All.X.glm.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 523 9
## Y 236 232
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 523
## 2 Y 9
## over50k.fctr.predict.All.X.glm.Y
## 1 236
## 2 232
## Reference
## Prediction N Y
## N 599 25
## Y 160 216
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 599
## 2 Y 25
## over50k.fctr.predict.All.X.glm.Y
## 1 160
## 2 216
## Reference
## Prediction N Y
## N 656 41
## Y 103 200
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 656
## 2 Y 41
## over50k.fctr.predict.All.X.glm.Y
## 1 103
## 2 200
## Reference
## Prediction N Y
## N 684 57
## Y 75 184
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 684
## 2 Y 57
## over50k.fctr.predict.All.X.glm.Y
## 1 75
## 2 184
## Reference
## Prediction N Y
## N 701 82
## Y 58 159
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 701
## 2 Y 82
## over50k.fctr.predict.All.X.glm.Y
## 1 58
## 2 159
## Reference
## Prediction N Y
## N 728 96
## Y 31 145
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 728
## 2 Y 96
## over50k.fctr.predict.All.X.glm.Y
## 1 31
## 2 145
## Reference
## Prediction N Y
## N 741 123
## Y 18 118
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 741
## 2 Y 123
## over50k.fctr.predict.All.X.glm.Y
## 1 18
## 2 118
## Reference
## Prediction N Y
## N 753 158
## Y 6 83
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 753
## 2 Y 158
## over50k.fctr.predict.All.X.glm.Y
## 1 6
## 2 83
## Reference
## Prediction N Y
## N 758 183
## Y 1 58
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 758
## 2 Y 183
## over50k.fctr.predict.All.X.glm.Y
## 1 1
## 2 58
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.All.X.glm.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.3883965
## 2 0.1 0.6544429
## 3 0.2 0.7001621
## 4 0.3 0.7352941
## 5 0.4 0.7360000
## 6 0.5 0.6943231
## 7 0.6 0.6954436
## 8 0.7 0.6259947
## 9 0.8 0.5030303
## 10 0.9 0.3866667
## 11 1.0 0.0000000## [1] "Classifier Probability Threshold: 0.4000 to maximize f.score.fit"
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 684
## 2 Y 57
## over50k.fctr.predict.All.X.glm.Y
## 1 75
## 2 184
## Reference
## Prediction N Y
## N 684 57
## Y 75 184
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 684
## 2 Y 57
## over50k.fctr.predict.All.X.glm.Y
## 1 75
## 2 184
## Prediction
## Reference N Y
## N 684 75
## Y 57 184
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.680000e-01 6.481520e-01 8.454346e-01 8.883720e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 5.968845e-18 1.389640e-01## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.All.X.glm.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 519 36
## Y 240 205
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 519
## 2 Y 36
## over50k.fctr.predict.All.X.glm.Y
## 1 240
## 2 205
## Reference
## Prediction N Y
## N 601 56
## Y 158 185
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 601
## 2 Y 56
## over50k.fctr.predict.All.X.glm.Y
## 1 158
## 2 185
## Reference
## Prediction N Y
## N 641 75
## Y 118 166
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 641
## 2 Y 75
## over50k.fctr.predict.All.X.glm.Y
## 1 118
## 2 166
## Reference
## Prediction N Y
## N 671 86
## Y 88 155
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 671
## 2 Y 86
## over50k.fctr.predict.All.X.glm.Y
## 1 88
## 2 155
## Reference
## Prediction N Y
## N 694 99
## Y 65 142
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 694
## 2 Y 99
## over50k.fctr.predict.All.X.glm.Y
## 1 65
## 2 142
## Reference
## Prediction N Y
## N 715 116
## Y 44 125
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 715
## 2 Y 116
## over50k.fctr.predict.All.X.glm.Y
## 1 44
## 2 125
## Reference
## Prediction N Y
## N 730 139
## Y 29 102
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 730
## 2 Y 139
## over50k.fctr.predict.All.X.glm.Y
## 1 29
## 2 102
## Reference
## Prediction N Y
## N 740 167
## Y 19 74
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 740
## 2 Y 167
## over50k.fctr.predict.All.X.glm.Y
## 1 19
## 2 74
## Reference
## Prediction N Y
## N 747 192
## Y 12 49
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 747
## 2 Y 192
## over50k.fctr.predict.All.X.glm.Y
## 1 12
## 2 49
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.All.X.glm.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.3883965
## 2 0.1 0.5976676
## 3 0.2 0.6335616
## 4 0.3 0.6323810
## 5 0.4 0.6404959
## 6 0.5 0.6339286
## 7 0.6 0.6097561
## 8 0.7 0.5483871
## 9 0.8 0.4431138
## 10 0.9 0.3245033
## 11 1.0 0.0000000
## [1] "Classifier Probability Threshold: 0.4000 to maximize f.score.OOB"
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 671
## 2 Y 86
## over50k.fctr.predict.All.X.glm.Y
## 1 88
## 2 155
## Reference
## Prediction N Y
## N 671 86
## Y 88 155
## over50k.fctr over50k.fctr.predict.All.X.glm.N
## 1 N 671
## 2 Y 86
## over50k.fctr.predict.All.X.glm.Y
## 1 88
## 2 155
## Prediction
## Reference N Y
## N 671 88
## Y 86 155
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.260000e-01 5.257227e-01 8.010531e-01 8.490099e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 1.789552e-07 9.395704e-01
## model_id model_method
## 1 All.X.glm glm
## feats
## 1 age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr, .rnorm
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 3.119 0.496
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.9325658 0.4 0.736 0.8299947
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.8454346 0.888372 0.5291816 0.835337
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.4 0.6404959 0.826
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB min.aic.fit
## 1 0.8010531 0.8490099 0.5257227 719.1617
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.05275524 0.1292318
## [1] "iterating over method:rpart"
## [1] "fitting model: All.X.rpart"
## [1] " indep_vars: age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr"
## + Fold1: cp=0.03734
## - Fold1: cp=0.03734
## + Fold2: cp=0.03734
## - Fold2: cp=0.03734
## + Fold3: cp=0.03734
## - Fold3: cp=0.03734
## Aggregating results
## Selecting tuning parameters
## Fitting cp = 0.0373 on full training set## Warning in myfit_mdl(model_id = paste0(model_id_pfx, ""), model_method =
## method, : model's bestTune found at an extreme of tuneGrid for parameter:
## cp
## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 1000
##
## CP nsplit rel error
## 1 0.07883817 0 1.0000000
## 2 0.04356846 2 0.8423237
## 3 0.03734440 5 0.6763485
##
## Variable importance
## maritalstatus.fctr Married-civ-spouse
## 28
## relationship.fctr Husband
## 25
## age
## 11
## capitalgain
## 10
## sex.fctr Female
## 9
## relationship.fctr Own-child
## 5
## maritalstatus.fctr Divorced
## 4
## occupation.fctr Prof-specialty
## 4
## occupation.fctr Exec-managerial
## 3
##
## Node number 1: 1000 observations, complexity param=0.07883817
## predicted class=N expected loss=0.241 P(node) =1
## class counts: 759 241
## probabilities: 0.759 0.241
## left son=2 (539 obs) right son=3 (461 obs)
## Primary splits:
## maritalstatus.fctr Married-civ-spouse < 0.5 to the left, improve=72.48758, (0 missing)
## relationship.fctr Husband < 0.5 to the left, improve=56.59330, (0 missing)
## capitalgain < 6897.5 to the left, improve=55.55498, (0 missing)
## age < 28.5 to the left, improve=33.75374, (0 missing)
## hoursperweek < 49.5 to the left, improve=21.28575, (0 missing)
## Surrogate splits:
## relationship.fctr Husband < 0.5 to the left, agree=0.947, adj=0.885, (0 split)
## sex.fctr Female < 0.5 to the right, agree=0.695, adj=0.338, (0 split)
## age < 31.5 to the left, agree=0.644, adj=0.228, (0 split)
## relationship.fctr Own-child < 0.5 to the right, agree=0.620, adj=0.176, (0 split)
## maritalstatus.fctr Divorced < 0.5 to the right, agree=0.609, adj=0.152, (0 split)
##
## Node number 2: 539 observations
## predicted class=N expected loss=0.06493506 P(node) =0.539
## class counts: 504 35
## probabilities: 0.935 0.065
##
## Node number 3: 461 observations, complexity param=0.07883817
## predicted class=N expected loss=0.4468547 P(node) =0.461
## class counts: 255 206
## probabilities: 0.553 0.447
## left son=6 (423 obs) right son=7 (38 obs)
## Primary splits:
## capitalgain < 5095.5 to the left, improve=25.342690, (0 missing)
## age < 28.5 to the left, improve=13.026120, (0 missing)
## occupation.fctr Prof-specialty < 0.5 to the left, improve=12.368610, (0 missing)
## education.fctr HS-grad < 0.5 to the right, improve=10.892940, (0 missing)
## hoursperweek < 49 to the left, improve= 8.044688, (0 missing)
##
## Node number 6: 423 observations, complexity param=0.04356846
## predicted class=N expected loss=0.3971631 P(node) =0.423
## class counts: 255 168
## probabilities: 0.603 0.397
## left son=12 (41 obs) right son=13 (382 obs)
## Primary splits:
## age < 28.5 to the left, improve=11.020560, (0 missing)
## education.fctr HS-grad < 0.5 to the right, improve= 8.436452, (0 missing)
## occupation.fctr Prof-specialty < 0.5 to the left, improve= 7.792584, (0 missing)
## capitalloss < 1813.5 to the left, improve= 7.227105, (0 missing)
## occupation.fctr Exec-managerial < 0.5 to the left, improve= 6.992129, (0 missing)
##
## Node number 7: 38 observations
## predicted class=Y expected loss=0 P(node) =0.038
## class counts: 0 38
## probabilities: 0.000 1.000
##
## Node number 12: 41 observations
## predicted class=N expected loss=0.04878049 P(node) =0.041
## class counts: 39 2
## probabilities: 0.951 0.049
##
## Node number 13: 382 observations, complexity param=0.04356846
## predicted class=N expected loss=0.434555 P(node) =0.382
## class counts: 216 166
## probabilities: 0.565 0.435
## left son=26 (315 obs) right son=27 (67 obs)
## Primary splits:
## occupation.fctr Exec-managerial < 0.5 to the left, improve=8.020381, (0 missing)
## education.fctr HS-grad < 0.5 to the right, improve=7.467051, (0 missing)
## occupation.fctr Prof-specialty < 0.5 to the left, improve=7.368971, (0 missing)
## occupation.fctr Other-service < 0.5 to the right, improve=6.306984, (0 missing)
## capitalloss < 1813.5 to the left, improve=5.929517, (0 missing)
##
## Node number 26: 315 observations, complexity param=0.04356846
## predicted class=N expected loss=0.3873016 P(node) =0.315
## class counts: 193 122
## probabilities: 0.613 0.387
## left son=52 (262 obs) right son=53 (53 obs)
## Primary splits:
## occupation.fctr Prof-specialty < 0.5 to the left, improve=10.862090, (0 missing)
## education.fctr HS-grad < 0.5 to the right, improve= 5.445683, (0 missing)
## workclass.fctr Federal-gov < 0.5 to the left, improve= 5.375036, (0 missing)
## occupation.fctr Other-service < 0.5 to the right, improve= 5.056941, (0 missing)
## occupation.fctr Sales < 0.5 to the left, improve= 4.800413, (0 missing)
## Surrogate splits:
## education.fctr Prof-school < 0.5 to the left, agree=0.851, adj=0.113, (0 split)
## education.fctr Masters < 0.5 to the left, agree=0.848, adj=0.094, (0 split)
## education.fctr Doctorate < 0.5 to the left, agree=0.848, adj=0.094, (0 split)
## capitalloss < 1894.5 to the left, agree=0.838, adj=0.038, (0 split)
##
## Node number 27: 67 observations
## predicted class=Y expected loss=0.3432836 P(node) =0.067
## class counts: 23 44
## probabilities: 0.343 0.657
##
## Node number 52: 262 observations
## predicted class=N expected loss=0.3282443 P(node) =0.262
## class counts: 176 86
## probabilities: 0.672 0.328
##
## Node number 53: 53 observations
## predicted class=Y expected loss=0.3207547 P(node) =0.053
## class counts: 17 36
## probabilities: 0.321 0.679
##
## n= 1000
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 1000 241 N (0.75900000 0.24100000)
## 2) maritalstatus.fctr Married-civ-spouse< 0.5 539 35 N (0.93506494 0.06493506) *
## 3) maritalstatus.fctr Married-civ-spouse>=0.5 461 206 N (0.55314534 0.44685466)
## 6) capitalgain< 5095.5 423 168 N (0.60283688 0.39716312)
## 12) age< 28.5 41 2 N (0.95121951 0.04878049) *
## 13) age>=28.5 382 166 N (0.56544503 0.43455497)
## 26) occupation.fctr Exec-managerial< 0.5 315 122 N (0.61269841 0.38730159)
## 52) occupation.fctr Prof-specialty< 0.5 262 86 N (0.67175573 0.32824427) *
## 53) occupation.fctr Prof-specialty>=0.5 53 17 Y (0.32075472 0.67924528) *
## 27) occupation.fctr Exec-managerial>=0.5 67 23 Y (0.34328358 0.65671642) *
## 7) capitalgain>=5095.5 38 0 Y (0.00000000 1.00000000) *
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.All.X.rpart.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 543 37
## Y 216 204
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 543
## 2 Y 37
## over50k.fctr.predict.All.X.rpart.Y
## 1 216
## 2 204
## Reference
## Prediction N Y
## N 543 37
## Y 216 204
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 543
## 2 Y 37
## over50k.fctr.predict.All.X.rpart.Y
## 1 216
## 2 204
## Reference
## Prediction N Y
## N 543 37
## Y 216 204
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 543
## 2 Y 37
## over50k.fctr.predict.All.X.rpart.Y
## 1 216
## 2 204
## Reference
## Prediction N Y
## N 719 123
## Y 40 118
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 719
## 2 Y 123
## over50k.fctr.predict.All.X.rpart.Y
## 1 40
## 2 118
## Reference
## Prediction N Y
## N 719 123
## Y 40 118
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 719
## 2 Y 123
## over50k.fctr.predict.All.X.rpart.Y
## 1 40
## 2 118
## Reference
## Prediction N Y
## N 719 123
## Y 40 118
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 719
## 2 Y 123
## over50k.fctr.predict.All.X.rpart.Y
## 1 40
## 2 118
## Reference
## Prediction N Y
## N 759 203
## Y 0 38
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 759
## 2 Y 203
## over50k.fctr.predict.All.X.rpart.Y
## 1 0
## 2 38
## Reference
## Prediction N Y
## N 759 203
## Y 0 38
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 759
## 2 Y 203
## over50k.fctr.predict.All.X.rpart.Y
## 1 0
## 2 38
## Reference
## Prediction N Y
## N 759 203
## Y 0 38
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 759
## 2 Y 203
## over50k.fctr.predict.All.X.rpart.Y
## 1 0
## 2 38
## Reference
## Prediction N Y
## N 759 203
## Y 0 38
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 759
## 2 Y 203
## over50k.fctr.predict.All.X.rpart.Y
## 1 0
## 2 38
## threshold f.score
## 1 0.0 0.3883965
## 2 0.1 0.6172466
## 3 0.2 0.6172466
## 4 0.3 0.6172466
## 5 0.4 0.5914787
## 6 0.5 0.5914787
## 7 0.6 0.5914787
## 8 0.7 0.2724014
## 9 0.8 0.2724014
## 10 0.9 0.2724014
## 11 1.0 0.2724014
## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.fit"
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 543
## 2 Y 37
## over50k.fctr.predict.All.X.rpart.Y
## 1 216
## 2 204
## Reference
## Prediction N Y
## N 543 37
## Y 216 204
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 543
## 2 Y 37
## over50k.fctr.predict.All.X.rpart.Y
## 1 216
## 2 204
## Prediction
## Reference N Y
## N 543 216
## Y 37 204
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.470000e-01 4.482729e-01 7.188527e-01 7.736848e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 8.225678e-01 4.525170e-29
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.All.X.rpart.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 539 44
## Y 220 197
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 539
## 2 Y 44
## over50k.fctr.predict.All.X.rpart.Y
## 1 220
## 2 197
## Reference
## Prediction N Y
## N 539 44
## Y 220 197
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 539
## 2 Y 44
## over50k.fctr.predict.All.X.rpart.Y
## 1 220
## 2 197
## Reference
## Prediction N Y
## N 539 44
## Y 220 197
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 539
## 2 Y 44
## over50k.fctr.predict.All.X.rpart.Y
## 1 220
## 2 197
## Reference
## Prediction N Y
## N 714 123
## Y 45 118
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 714
## 2 Y 123
## over50k.fctr.predict.All.X.rpart.Y
## 1 45
## 2 118
## Reference
## Prediction N Y
## N 714 123
## Y 45 118
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 714
## 2 Y 123
## over50k.fctr.predict.All.X.rpart.Y
## 1 45
## 2 118
## Reference
## Prediction N Y
## N 714 123
## Y 45 118
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 714
## 2 Y 123
## over50k.fctr.predict.All.X.rpart.Y
## 1 45
## 2 118
## Reference
## Prediction N Y
## N 759 210
## Y 0 31
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 759
## 2 Y 210
## over50k.fctr.predict.All.X.rpart.Y
## 1 0
## 2 31
## Reference
## Prediction N Y
## N 759 210
## Y 0 31
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 759
## 2 Y 210
## over50k.fctr.predict.All.X.rpart.Y
## 1 0
## 2 31
## Reference
## Prediction N Y
## N 759 210
## Y 0 31
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 759
## 2 Y 210
## over50k.fctr.predict.All.X.rpart.Y
## 1 0
## 2 31
## Reference
## Prediction N Y
## N 759 210
## Y 0 31
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 759
## 2 Y 210
## over50k.fctr.predict.All.X.rpart.Y
## 1 0
## 2 31
## threshold f.score
## 1 0.0 0.3883965
## 2 0.1 0.5987842
## 3 0.2 0.5987842
## 4 0.3 0.5987842
## 5 0.4 0.5841584
## 6 0.5 0.5841584
## 7 0.6 0.5841584
## 8 0.7 0.2279412
## 9 0.8 0.2279412
## 10 0.9 0.2279412
## 11 1.0 0.2279412
## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.OOB"
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 539
## 2 Y 44
## over50k.fctr.predict.All.X.rpart.Y
## 1 220
## 2 197
## Reference
## Prediction N Y
## N 539 44
## Y 220 197
## over50k.fctr over50k.fctr.predict.All.X.rpart.N
## 1 N 539
## 2 Y 44
## over50k.fctr.predict.All.X.rpart.Y
## 1 220
## 2 197
## Prediction
## Reference N Y
## N 539 220
## Y 44 197
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 7.360000e-01 4.223271e-01 7.075104e-01 7.630926e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 9.577710e-01 4.743896e-27
## model_id model_method
## 1 All.X.rpart rpart
## feats
## 1 age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 3 2.392 0.247
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.8336586 0.3 0.6172466 0.8020056
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.7188527 0.7736848 0.3405157 0.8063651
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.3 0.5987842 0.736
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.7075104 0.7630926 0.4223271
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.005703043 0.1004617
## [1] "fitting model: All.X.cp.0.rpart"
## [1] " indep_vars: age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr"
## Fitting cp = 0 on full training set
## Call:
## rpart(formula = .outcome ~ ., control = list(minsplit = 20, minbucket = 7,
## cp = 0, maxcompete = 4, maxsurrogate = 5, usesurrogate = 2,
## surrogatestyle = 0, maxdepth = 30, xval = 0))
## n= 1000
##
## CP nsplit rel error
## 1 0.078838174 0 1.0000000
## 2 0.043568465 2 0.8423237
## 3 0.037344398 5 0.6763485
## 4 0.020746888 7 0.6016598
## 5 0.016597510 8 0.5809129
## 6 0.012448133 9 0.5643154
## 7 0.004149378 10 0.5518672
## 8 0.002074689 17 0.5186722
## 9 0.000000000 23 0.5062241
##
## Variable importance
## maritalstatus.fctr Married-civ-spouse
## 23
## relationship.fctr Husband
## 20
## capitalgain
## 13
## age
## 10
## sex.fctr Female
## 8
## relationship.fctr Own-child
## 4
## maritalstatus.fctr Divorced
## 3
## occupation.fctr Prof-specialty
## 3
## occupation.fctr Sales
## 3
## occupation.fctr Exec-managerial
## 3
## hoursperweek
## 2
## occupation.fctr Tech-support
## 2
## capitalloss
## 1
## workclass.fctr Self-emp-not-inc
## 1
## education.fctr HS-grad
## 1
## occupation.fctr Other-service
## 1
## occupation.fctr Craft-repair
## 1
## workclass.fctr Self-emp-inc
## 1
##
## Node number 1: 1000 observations, complexity param=0.07883817
## predicted class=N expected loss=0.241 P(node) =1
## class counts: 759 241
## probabilities: 0.759 0.241
## left son=2 (539 obs) right son=3 (461 obs)
## Primary splits:
## maritalstatus.fctr Married-civ-spouse < 0.5 to the left, improve=72.48758, (0 missing)
## relationship.fctr Husband < 0.5 to the left, improve=56.59330, (0 missing)
## capitalgain < 6897.5 to the left, improve=55.55498, (0 missing)
## age < 28.5 to the left, improve=33.75374, (0 missing)
## hoursperweek < 49.5 to the left, improve=21.28575, (0 missing)
## Surrogate splits:
## relationship.fctr Husband < 0.5 to the left, agree=0.947, adj=0.885, (0 split)
## sex.fctr Female < 0.5 to the right, agree=0.695, adj=0.338, (0 split)
## age < 31.5 to the left, agree=0.644, adj=0.228, (0 split)
## relationship.fctr Own-child < 0.5 to the right, agree=0.620, adj=0.176, (0 split)
## maritalstatus.fctr Divorced < 0.5 to the right, agree=0.609, adj=0.152, (0 split)
##
## Node number 2: 539 observations, complexity param=0.0373444
## predicted class=N expected loss=0.06493506 P(node) =0.539
## class counts: 504 35
## probabilities: 0.935 0.065
## left son=4 (526 obs) right son=5 (13 obs)
## Primary splits:
## capitalgain < 4718.5 to the left, improve=16.260040, (0 missing)
## hoursperweek < 44.5 to the left, improve= 4.598427, (0 missing)
## occupation.fctr Exec-managerial < 0.5 to the left, improve= 3.861460, (0 missing)
## age < 29.5 to the left, improve= 2.470557, (0 missing)
## education.fctr Masters < 0.5 to the left, improve= 1.310401, (0 missing)
##
## Node number 3: 461 observations, complexity param=0.07883817
## predicted class=N expected loss=0.4468547 P(node) =0.461
## class counts: 255 206
## probabilities: 0.553 0.447
## left son=6 (423 obs) right son=7 (38 obs)
## Primary splits:
## capitalgain < 5095.5 to the left, improve=25.342690, (0 missing)
## age < 28.5 to the left, improve=13.026120, (0 missing)
## occupation.fctr Prof-specialty < 0.5 to the left, improve=12.368610, (0 missing)
## education.fctr HS-grad < 0.5 to the right, improve=10.892940, (0 missing)
## hoursperweek < 49 to the left, improve= 8.044688, (0 missing)
##
## Node number 4: 526 observations
## predicted class=N expected loss=0.04562738 P(node) =0.526
## class counts: 502 24
## probabilities: 0.954 0.046
##
## Node number 5: 13 observations
## predicted class=Y expected loss=0.1538462 P(node) =0.013
## class counts: 2 11
## probabilities: 0.154 0.846
##
## Node number 6: 423 observations, complexity param=0.04356846
## predicted class=N expected loss=0.3971631 P(node) =0.423
## class counts: 255 168
## probabilities: 0.603 0.397
## left son=12 (41 obs) right son=13 (382 obs)
## Primary splits:
## age < 28.5 to the left, improve=11.020560, (0 missing)
## education.fctr HS-grad < 0.5 to the right, improve= 8.436452, (0 missing)
## occupation.fctr Prof-specialty < 0.5 to the left, improve= 7.792584, (0 missing)
## capitalloss < 1813.5 to the left, improve= 7.227105, (0 missing)
## occupation.fctr Exec-managerial < 0.5 to the left, improve= 6.992129, (0 missing)
##
## Node number 7: 38 observations
## predicted class=Y expected loss=0 P(node) =0.038
## class counts: 0 38
## probabilities: 0.000 1.000
##
## Node number 12: 41 observations
## predicted class=N expected loss=0.04878049 P(node) =0.041
## class counts: 39 2
## probabilities: 0.951 0.049
##
## Node number 13: 382 observations, complexity param=0.04356846
## predicted class=N expected loss=0.434555 P(node) =0.382
## class counts: 216 166
## probabilities: 0.565 0.435
## left son=26 (315 obs) right son=27 (67 obs)
## Primary splits:
## occupation.fctr Exec-managerial < 0.5 to the left, improve=8.020381, (0 missing)
## education.fctr HS-grad < 0.5 to the right, improve=7.467051, (0 missing)
## occupation.fctr Prof-specialty < 0.5 to the left, improve=7.368971, (0 missing)
## occupation.fctr Other-service < 0.5 to the right, improve=6.306984, (0 missing)
## capitalloss < 1813.5 to the left, improve=5.929517, (0 missing)
##
## Node number 26: 315 observations, complexity param=0.04356846
## predicted class=N expected loss=0.3873016 P(node) =0.315
## class counts: 193 122
## probabilities: 0.613 0.387
## left son=52 (262 obs) right son=53 (53 obs)
## Primary splits:
## occupation.fctr Prof-specialty < 0.5 to the left, improve=10.862090, (0 missing)
## education.fctr HS-grad < 0.5 to the right, improve= 5.445683, (0 missing)
## workclass.fctr Federal-gov < 0.5 to the left, improve= 5.375036, (0 missing)
## occupation.fctr Other-service < 0.5 to the right, improve= 5.056941, (0 missing)
## occupation.fctr Sales < 0.5 to the left, improve= 4.800413, (0 missing)
## Surrogate splits:
## education.fctr Prof-school < 0.5 to the left, agree=0.851, adj=0.113, (0 split)
## education.fctr Masters < 0.5 to the left, agree=0.848, adj=0.094, (0 split)
## education.fctr Doctorate < 0.5 to the left, agree=0.848, adj=0.094, (0 split)
## capitalloss < 1894.5 to the left, agree=0.838, adj=0.038, (0 split)
##
## Node number 27: 67 observations, complexity param=0.01244813
## predicted class=Y expected loss=0.3432836 P(node) =0.067
## class counts: 23 44
## probabilities: 0.343 0.657
## left son=54 (7 obs) right son=55 (60 obs)
## Primary splits:
## workclass.fctr Self-emp-not-inc < 0.5 to the right, improve=2.1518120, (0 missing)
## capitalloss < 1649 to the left, improve=2.1411590, (0 missing)
## education.fctr HS-grad < 0.5 to the right, improve=1.4004450, (0 missing)
## workclass.fctr Private < 0.5 to the left, improve=0.8810273, (0 missing)
## hoursperweek < 33.5 to the left, improve=0.8137171, (0 missing)
##
## Node number 52: 262 observations, complexity param=0.0373444
## predicted class=N expected loss=0.3282443 P(node) =0.262
## class counts: 176 86
## probabilities: 0.672 0.328
## left son=104 (215 obs) right son=105 (47 obs)
## Primary splits:
## occupation.fctr Sales < 0.5 to the left, improve=8.196710, (0 missing)
## occupation.fctr Tech-support < 0.5 to the left, improve=3.766842, (0 missing)
## occupation.fctr Other-service < 0.5 to the right, improve=3.672066, (0 missing)
## hoursperweek < 33.5 to the left, improve=2.574169, (0 missing)
## workclass.fctr Self-emp-inc < 0.5 to the left, improve=2.255608, (0 missing)
## Surrogate splits:
## workclass.fctr Self-emp-inc < 0.5 to the left, agree=0.840, adj=0.106, (0 split)
## hoursperweek < 77.5 to the left, agree=0.832, adj=0.064, (0 split)
##
## Node number 53: 53 observations, complexity param=0.002074689
## predicted class=Y expected loss=0.3207547 P(node) =0.053
## class counts: 17 36
## probabilities: 0.321 0.679
## left son=106 (45 obs) right son=107 (8 obs)
## Primary splits:
## capitalloss < 951 to the left, improve=1.9387840, (0 missing)
## hoursperweek < 42.5 to the left, improve=1.4521110, (0 missing)
## age < 34.5 to the left, improve=1.1953500, (0 missing)
## relationship.fctr Wife < 0.5 to the left, improve=0.9529255, (0 missing)
## sex.fctr Female < 0.5 to the left, improve=0.9529255, (0 missing)
##
## Node number 54: 7 observations
## predicted class=N expected loss=0.2857143 P(node) =0.007
## class counts: 5 2
## probabilities: 0.714 0.286
##
## Node number 55: 60 observations, complexity param=0.002074689
## predicted class=Y expected loss=0.3 P(node) =0.06
## class counts: 18 42
## probabilities: 0.300 0.700
## left son=110 (52 obs) right son=111 (8 obs)
## Primary splits:
## capitalloss < 1649 to the left, improve=1.6615380, (0 missing)
## hoursperweek < 36.5 to the left, improve=1.1676550, (0 missing)
## education.fctr HS-grad < 0.5 to the right, improve=1.0730160, (0 missing)
## age < 55 to the right, improve=0.9600000, (0 missing)
## sex.fctr Female < 0.5 to the right, improve=0.2619946, (0 missing)
##
## Node number 104: 215 observations, complexity param=0.02074689
## predicted class=N expected loss=0.2697674 P(node) =0.215
## class counts: 157 58
## probabilities: 0.730 0.270
## left son=208 (206 obs) right son=209 (9 obs)
## Primary splits:
## occupation.fctr Tech-support < 0.5 to the left, improve=4.848293, (0 missing)
## occupation.fctr Other-service < 0.5 to the right, improve=2.516022, (0 missing)
## hoursperweek < 33.5 to the left, improve=2.199854, (0 missing)
## occupation.fctr Farming-fishing < 0.5 to the right, improve=1.526489, (0 missing)
## race.fctr Black < 0.5 to the left, improve=1.432618, (0 missing)
##
## Node number 105: 47 observations, complexity param=0.01659751
## predicted class=Y expected loss=0.4042553 P(node) =0.047
## class counts: 19 28
## probabilities: 0.404 0.596
## left son=210 (12 obs) right son=211 (35 obs)
## Primary splits:
## education.fctr HS-grad < 0.5 to the right, improve=2.2192500, (0 missing)
## workclass.fctr Self-emp-inc < 0.5 to the left, improve=1.9131520, (0 missing)
## workclass.fctr Private < 0.5 to the right, improve=0.5603758, (0 missing)
## hoursperweek < 38 to the left, improve=0.4597264, (0 missing)
## age < 38.5 to the left, improve=0.4447495, (0 missing)
## Surrogate splits:
## age < 62.5 to the right, agree=0.766, adj=0.083, (0 split)
## workclass.fctr Self-emp-not-inc < 0.5 to the right, agree=0.766, adj=0.083, (0 split)
##
## Node number 106: 45 observations, complexity param=0.002074689
## predicted class=Y expected loss=0.3777778 P(node) =0.045
## class counts: 17 28
## probabilities: 0.378 0.622
## left son=212 (27 obs) right son=213 (18 obs)
## Primary splits:
## hoursperweek < 42.5 to the left, improve=1.4518520, (0 missing)
## relationship.fctr Wife < 0.5 to the left, improve=1.2433930, (0 missing)
## sex.fctr Female < 0.5 to the left, improve=1.2433930, (0 missing)
## age < 34.5 to the left, improve=1.1893390, (0 missing)
## relationship.fctr Husband < 0.5 to the right, improve=0.5444444, (0 missing)
## Surrogate splits:
## age < 60.5 to the left, agree=0.644, adj=0.111, (0 split)
## workclass.fctr Self-emp-inc < 0.5 to the left, agree=0.644, adj=0.111, (0 split)
## workclass.fctr Federal-gov < 0.5 to the left, agree=0.622, adj=0.056, (0 split)
## education.fctr Prof-school < 0.5 to the left, agree=0.622, adj=0.056, (0 split)
##
## Node number 107: 8 observations
## predicted class=Y expected loss=0 P(node) =0.008
## class counts: 0 8
## probabilities: 0.000 1.000
##
## Node number 110: 52 observations, complexity param=0.002074689
## predicted class=Y expected loss=0.3461538 P(node) =0.052
## class counts: 18 34
## probabilities: 0.346 0.654
## left son=220 (15 obs) right son=221 (37 obs)
## Primary splits:
## education.fctr HS-grad < 0.5 to the right, improve=1.4772000, (0 missing)
## hoursperweek < 51.5 to the left, improve=1.2820510, (0 missing)
## age < 55 to the right, improve=0.9544822, (0 missing)
## workclass.fctr Private < 0.5 to the left, improve=0.1884615, (0 missing)
## education.fctr Masters < 0.5 to the left, improve=0.1748252, (0 missing)
## Surrogate splits:
## age < 55 to the right, agree=0.808, adj=0.333, (0 split)
## hoursperweek < 22 to the left, agree=0.750, adj=0.133, (0 split)
##
## Node number 111: 8 observations
## predicted class=Y expected loss=0 P(node) =0.008
## class counts: 0 8
## probabilities: 0.000 1.000
##
## Node number 208: 206 observations, complexity param=0.004149378
## predicted class=N expected loss=0.2475728 P(node) =0.206
## class counts: 155 51
## probabilities: 0.752 0.248
## left son=416 (16 obs) right son=417 (190 obs)
## Primary splits:
## occupation.fctr Other-service < 0.5 to the right, improve=2.126520, (0 missing)
## hoursperweek < 33.5 to the left, improve=1.733043, (0 missing)
## workclass.fctr Self-emp-not-inc < 0.5 to the right, improve=1.335836, (0 missing)
## occupation.fctr Farming-fishing < 0.5 to the right, improve=1.288389, (0 missing)
## race.fctr Black < 0.5 to the left, improve=1.259816, (0 missing)
## Surrogate splits:
## capitalgain < 3941 to the right, agree=0.927, adj=0.062, (0 split)
##
## Node number 209: 9 observations
## predicted class=Y expected loss=0.2222222 P(node) =0.009
## class counts: 2 7
## probabilities: 0.222 0.778
##
## Node number 210: 12 observations
## predicted class=N expected loss=0.3333333 P(node) =0.012
## class counts: 8 4
## probabilities: 0.667 0.333
##
## Node number 211: 35 observations
## predicted class=Y expected loss=0.3142857 P(node) =0.035
## class counts: 11 24
## probabilities: 0.314 0.686
##
## Node number 212: 27 observations, complexity param=0.002074689
## predicted class=Y expected loss=0.4814815 P(node) =0.027
## class counts: 13 14
## probabilities: 0.481 0.519
## left son=424 (7 obs) right son=425 (20 obs)
## Primary splits:
## age < 49 to the right, improve=0.15291010, (0 missing)
## education.fctr Masters < 0.5 to the left, improve=0.03703704, (0 missing)
## workclass.fctr Private < 0.5 to the left, improve=0.02693603, (0 missing)
## workclass.fctr Local-gov < 0.5 to the left, improve=0.02693603, (0 missing)
## Surrogate splits:
## workclass.fctr Self-emp-not-inc < 0.5 to the right, agree=0.778, adj=0.143, (0 split)
##
## Node number 213: 18 observations
## predicted class=Y expected loss=0.2222222 P(node) =0.018
## class counts: 4 14
## probabilities: 0.222 0.778
##
## Node number 220: 15 observations
## predicted class=N expected loss=0.4666667 P(node) =0.015
## class counts: 8 7
## probabilities: 0.533 0.467
##
## Node number 221: 37 observations
## predicted class=Y expected loss=0.2702703 P(node) =0.037
## class counts: 10 27
## probabilities: 0.270 0.730
##
## Node number 416: 16 observations
## predicted class=N expected loss=0 P(node) =0.016
## class counts: 16 0
## probabilities: 1.000 0.000
##
## Node number 417: 190 observations, complexity param=0.004149378
## predicted class=N expected loss=0.2684211 P(node) =0.19
## class counts: 139 51
## probabilities: 0.732 0.268
## left son=834 (22 obs) right son=835 (168 obs)
## Primary splits:
## workclass.fctr Self-emp-not-inc < 0.5 to the right, improve=1.568022, (0 missing)
## occupation.fctr Farming-fishing < 0.5 to the right, improve=1.521053, (0 missing)
## hoursperweek < 33.5 to the left, improve=1.416206, (0 missing)
## capitalloss < 629 to the right, improve=1.361384, (0 missing)
## race.fctr Black < 0.5 to the left, improve=1.232164, (0 missing)
##
## Node number 424: 7 observations
## predicted class=N expected loss=0.4285714 P(node) =0.007
## class counts: 4 3
## probabilities: 0.571 0.429
##
## Node number 425: 20 observations, complexity param=0.002074689
## predicted class=Y expected loss=0.45 P(node) =0.02
## class counts: 9 11
## probabilities: 0.450 0.550
## left son=850 (9 obs) right son=851 (11 obs)
## Primary splits:
## age < 38 to the left, improve=0.364646500, (0 missing)
## workclass.fctr Local-gov < 0.5 to the right, improve=0.066666670, (0 missing)
## workclass.fctr Private < 0.5 to the left, improve=0.001010101, (0 missing)
## Surrogate splits:
## workclass.fctr Private < 0.5 to the left, agree=0.70, adj=0.333, (0 split)
## hoursperweek < 37.5 to the left, agree=0.65, adj=0.222, (0 split)
## workclass.fctr Local-gov < 0.5 to the right, agree=0.65, adj=0.222, (0 split)
## education.fctr HS-grad < 0.5 to the right, agree=0.65, adj=0.222, (0 split)
## relationship.fctr Husband < 0.5 to the left, agree=0.60, adj=0.111, (0 split)
##
## Node number 834: 22 observations
## predicted class=N expected loss=0.09090909 P(node) =0.022
## class counts: 20 2
## probabilities: 0.909 0.091
##
## Node number 835: 168 observations, complexity param=0.004149378
## predicted class=N expected loss=0.2916667 P(node) =0.168
## class counts: 119 49
## probabilities: 0.708 0.292
## left son=1670 (27 obs) right son=1671 (141 obs)
## Primary splits:
## age < 60 to the right, improve=1.3252560, (0 missing)
## capitalloss < 629 to the right, improve=1.2427540, (0 missing)
## occupation.fctr Protective-serv < 0.5 to the right, improve=1.2427540, (0 missing)
## education.fctr 7th-8th < 0.5 to the right, improve=0.9488033, (0 missing)
## race.fctr Black < 0.5 to the left, improve=0.9411111, (0 missing)
## Surrogate splits:
## hoursperweek < 18 to the left, agree=0.869, adj=0.185, (0 split)
## capitalloss < 1706 to the right, agree=0.863, adj=0.148, (0 split)
## workclass.fctr ? < 0.5 to the right, agree=0.857, adj=0.111, (0 split)
## occupation.fctr ? < 0.5 to the right, agree=0.857, adj=0.111, (0 split)
## capitalgain < 3467.5 to the right, agree=0.845, adj=0.037, (0 split)
##
## Node number 850: 9 observations
## predicted class=N expected loss=0.4444444 P(node) =0.009
## class counts: 5 4
## probabilities: 0.556 0.444
##
## Node number 851: 11 observations
## predicted class=Y expected loss=0.3636364 P(node) =0.011
## class counts: 4 7
## probabilities: 0.364 0.636
##
## Node number 1670: 27 observations
## predicted class=N expected loss=0.1481481 P(node) =0.027
## class counts: 23 4
## probabilities: 0.852 0.148
##
## Node number 1671: 141 observations, complexity param=0.004149378
## predicted class=N expected loss=0.3191489 P(node) =0.141
## class counts: 96 45
## probabilities: 0.681 0.319
## left son=3342 (93 obs) right son=3343 (48 obs)
## Primary splits:
## age < 44.5 to the left, improve=1.3841230, (0 missing)
## race.fctr Black < 0.5 to the left, improve=1.1805960, (0 missing)
## education.fctr 7th-8th < 0.5 to the right, improve=0.6393777, (0 missing)
## hoursperweek < 52.5 to the right, improve=0.4765957, (0 missing)
## occupation.fctr Craft-repair < 0.5 to the right, improve=0.4637811, (0 missing)
## Surrogate splits:
## workclass.fctr Federal-gov < 0.5 to the left, agree=0.688, adj=0.083, (0 split)
## capitalgain < 2841.5 to the left, agree=0.674, adj=0.042, (0 split)
## education.fctr 10th < 0.5 to the left, agree=0.667, adj=0.021, (0 split)
##
## Node number 3342: 93 observations
## predicted class=N expected loss=0.2688172 P(node) =0.093
## class counts: 68 25
## probabilities: 0.731 0.269
##
## Node number 3343: 48 observations, complexity param=0.004149378
## predicted class=N expected loss=0.4166667 P(node) =0.048
## class counts: 28 20
## probabilities: 0.583 0.417
## left son=6686 (15 obs) right son=6687 (33 obs)
## Primary splits:
## occupation.fctr Craft-repair < 0.5 to the right, improve=2.0484850, (0 missing)
## hoursperweek < 51 to the right, improve=1.2288040, (0 missing)
## age < 52.5 to the right, improve=1.1111110, (0 missing)
## occupation.fctr Machine-op-inspct < 0.5 to the left, improve=0.4733825, (0 missing)
## education.fctr Some-college < 0.5 to the right, improve=0.2810685, (0 missing)
##
## Node number 6686: 15 observations
## predicted class=N expected loss=0.2 P(node) =0.015
## class counts: 12 3
## probabilities: 0.800 0.200
##
## Node number 6687: 33 observations, complexity param=0.004149378
## predicted class=Y expected loss=0.4848485 P(node) =0.033
## class counts: 16 17
## probabilities: 0.485 0.515
## left son=13374 (7 obs) right son=13375 (26 obs)
## Primary splits:
## hoursperweek < 51 to the right, improve=2.46287000, (0 missing)
## age < 52.5 to the right, improve=1.84638700, (0 missing)
## occupation.fctr Transport-moving < 0.5 to the right, improve=0.81818180, (0 missing)
## race.fctr Black < 0.5 to the right, improve=0.13320010, (0 missing)
## occupation.fctr Machine-op-inspct < 0.5 to the left, improve=0.03030303, (0 missing)
## Surrogate splits:
## workclass.fctr Self-emp-inc < 0.5 to the right, agree=0.848, adj=0.286, (0 split)
## education.fctr Some-college < 0.5 to the right, agree=0.818, adj=0.143, (0 split)
##
## Node number 13374: 7 observations
## predicted class=N expected loss=0.1428571 P(node) =0.007
## class counts: 6 1
## probabilities: 0.857 0.143
##
## Node number 13375: 26 observations, complexity param=0.004149378
## predicted class=Y expected loss=0.3846154 P(node) =0.026
## class counts: 10 16
## probabilities: 0.385 0.615
## left son=26750 (10 obs) right son=26751 (16 obs)
## Primary splits:
## age < 52.5 to the right, improve=1.507692000, (0 missing)
## occupation.fctr Transport-moving < 0.5 to the right, improve=1.335470000, (0 missing)
## occupation.fctr Machine-op-inspct < 0.5 to the right, improve=0.432692300, (0 missing)
## education.fctr HS-grad < 0.5 to the right, improve=0.007692308, (0 missing)
## Surrogate splits:
## capitalgain < 1551.5 to the right, agree=0.731, adj=0.3, (0 split)
## education.fctr 10th < 0.5 to the right, agree=0.731, adj=0.3, (0 split)
##
## Node number 26750: 10 observations
## predicted class=N expected loss=0.4 P(node) =0.01
## class counts: 6 4
## probabilities: 0.600 0.400
##
## Node number 26751: 16 observations
## predicted class=Y expected loss=0.25 P(node) =0.016
## class counts: 4 12
## probabilities: 0.250 0.750
##
## n= 1000
##
## node), split, n, loss, yval, (yprob)
## * denotes terminal node
##
## 1) root 1000 241 N (0.75900000 0.24100000)
## 2) maritalstatus.fctr Married-civ-spouse< 0.5 539 35 N (0.93506494 0.06493506)
## 4) capitalgain< 4718.5 526 24 N (0.95437262 0.04562738) *
## 5) capitalgain>=4718.5 13 2 Y (0.15384615 0.84615385) *
## 3) maritalstatus.fctr Married-civ-spouse>=0.5 461 206 N (0.55314534 0.44685466)
## 6) capitalgain< 5095.5 423 168 N (0.60283688 0.39716312)
## 12) age< 28.5 41 2 N (0.95121951 0.04878049) *
## 13) age>=28.5 382 166 N (0.56544503 0.43455497)
## 26) occupation.fctr Exec-managerial< 0.5 315 122 N (0.61269841 0.38730159)
## 52) occupation.fctr Prof-specialty< 0.5 262 86 N (0.67175573 0.32824427)
## 104) occupation.fctr Sales< 0.5 215 58 N (0.73023256 0.26976744)
## 208) occupation.fctr Tech-support< 0.5 206 51 N (0.75242718 0.24757282)
## 416) occupation.fctr Other-service>=0.5 16 0 N (1.00000000 0.00000000) *
## 417) occupation.fctr Other-service< 0.5 190 51 N (0.73157895 0.26842105)
## 834) workclass.fctr Self-emp-not-inc>=0.5 22 2 N (0.90909091 0.09090909) *
## 835) workclass.fctr Self-emp-not-inc< 0.5 168 49 N (0.70833333 0.29166667)
## 1670) age>=60 27 4 N (0.85185185 0.14814815) *
## 1671) age< 60 141 45 N (0.68085106 0.31914894)
## 3342) age< 44.5 93 25 N (0.73118280 0.26881720) *
## 3343) age>=44.5 48 20 N (0.58333333 0.41666667)
## 6686) occupation.fctr Craft-repair>=0.5 15 3 N (0.80000000 0.20000000) *
## 6687) occupation.fctr Craft-repair< 0.5 33 16 Y (0.48484848 0.51515152)
## 13374) hoursperweek>=51 7 1 N (0.85714286 0.14285714) *
## 13375) hoursperweek< 51 26 10 Y (0.38461538 0.61538462)
## 26750) age>=52.5 10 4 N (0.60000000 0.40000000) *
## 26751) age< 52.5 16 4 Y (0.25000000 0.75000000) *
## 209) occupation.fctr Tech-support>=0.5 9 2 Y (0.22222222 0.77777778) *
## 105) occupation.fctr Sales>=0.5 47 19 Y (0.40425532 0.59574468)
## 210) education.fctr HS-grad>=0.5 12 4 N (0.66666667 0.33333333) *
## 211) education.fctr HS-grad< 0.5 35 11 Y (0.31428571 0.68571429) *
## 53) occupation.fctr Prof-specialty>=0.5 53 17 Y (0.32075472 0.67924528)
## 106) capitalloss< 951 45 17 Y (0.37777778 0.62222222)
## 212) hoursperweek< 42.5 27 13 Y (0.48148148 0.51851852)
## 424) age>=49 7 3 N (0.57142857 0.42857143) *
## 425) age< 49 20 9 Y (0.45000000 0.55000000)
## 850) age< 38 9 4 N (0.55555556 0.44444444) *
## 851) age>=38 11 4 Y (0.36363636 0.63636364) *
## 213) hoursperweek>=42.5 18 4 Y (0.22222222 0.77777778) *
## 107) capitalloss>=951 8 0 Y (0.00000000 1.00000000) *
## 27) occupation.fctr Exec-managerial>=0.5 67 23 Y (0.34328358 0.65671642)
## 54) workclass.fctr Self-emp-not-inc>=0.5 7 2 N (0.71428571 0.28571429) *
## 55) workclass.fctr Self-emp-not-inc< 0.5 60 18 Y (0.30000000 0.70000000)
## 110) capitalloss< 1649 52 18 Y (0.34615385 0.65384615)
## 220) education.fctr HS-grad>=0.5 15 7 N (0.53333333 0.46666667) *
## 221) education.fctr HS-grad< 0.5 37 10 Y (0.27027027 0.72972973) *
## 111) capitalloss>=1649 8 0 Y (0.00000000 1.00000000) *
## 7) capitalgain>=5095.5 38 0 Y (0.00000000 1.00000000) *
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 577 28
## Y 182 213
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 577
## 2 Y 28
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 182
## 2 213
## Reference
## Prediction N Y
## N 606 33
## Y 153 208
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 606
## 2 Y 33
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 153
## 2 208
## Reference
## Prediction N Y
## N 691 63
## Y 68 178
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 691
## 2 Y 63
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 68
## 2 178
## Reference
## Prediction N Y
## N 699 67
## Y 60 174
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 699
## 2 Y 67
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 60
## 2 174
## Reference
## Prediction N Y
## N 722 85
## Y 37 156
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 722
## 2 Y 85
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 37
## 2 156
## Reference
## Prediction N Y
## N 722 85
## Y 37 156
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 722
## 2 Y 85
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 37
## 2 156
## Reference
## Prediction N Y
## N 737 116
## Y 22 125
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 737
## 2 Y 116
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 22
## 2 125
## Reference
## Prediction N Y
## N 757 176
## Y 2 65
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 757
## 2 Y 176
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 2
## 2 65
## Reference
## Prediction N Y
## N 759 187
## Y 0 54
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 759
## 2 Y 187
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 0
## 2 54
## Reference
## Prediction N Y
## N 759 187
## Y 0 54
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 759
## 2 Y 187
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 0
## 2 54
## threshold f.score
## 1 0.0 0.3883965
## 2 0.1 0.6698113
## 3 0.2 0.6910299
## 4 0.3 0.7310062
## 5 0.4 0.7326316
## 6 0.5 0.7188940
## 7 0.6 0.7188940
## 8 0.7 0.6443299
## 9 0.8 0.4220779
## 10 0.9 0.3661017
## 11 1.0 0.3661017
## [1] "Classifier Probability Threshold: 0.4000 to maximize f.score.fit"
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 699
## 2 Y 67
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 60
## 2 174
## Reference
## Prediction N Y
## N 699 67
## Y 60 174
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 699
## 2 Y 67
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 60
## 2 174
## Prediction
## Reference N Y
## N 699 60
## Y 67 174
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.730000e-01 6.493766e-01 8.507600e-01 8.930157e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 1.318095e-19 5.944394e-01
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 564 39
## Y 195 202
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 564
## 2 Y 39
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 195
## 2 202
## Reference
## Prediction N Y
## N 586 48
## Y 173 193
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 586
## 2 Y 48
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 173
## 2 193
## Reference
## Prediction N Y
## N 664 83
## Y 95 158
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 664
## 2 Y 83
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 95
## 2 158
## Reference
## Prediction N Y
## N 673 88
## Y 86 153
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 673
## 2 Y 88
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 86
## 2 153
## Reference
## Prediction N Y
## N 698 113
## Y 61 128
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 698
## 2 Y 113
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 61
## 2 128
## Reference
## Prediction N Y
## N 698 113
## Y 61 128
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 698
## 2 Y 113
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 61
## 2 128
## Reference
## Prediction N Y
## N 718 134
## Y 41 107
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 718
## 2 Y 134
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 41
## 2 107
## Reference
## Prediction N Y
## N 755 188
## Y 4 53
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 755
## 2 Y 188
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 4
## 2 53
## Reference
## Prediction N Y
## N 758 199
## Y 1 42
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 758
## 2 Y 199
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 1
## 2 42
## Reference
## Prediction N Y
## N 758 199
## Y 1 42
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 758
## 2 Y 199
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 1
## 2 42
## threshold f.score
## 1 0.0 0.3883965
## 2 0.1 0.6332288
## 3 0.2 0.6359143
## 4 0.3 0.6396761
## 5 0.4 0.6375000
## 6 0.5 0.5953488
## 7 0.6 0.5953488
## 8 0.7 0.5501285
## 9 0.8 0.3557047
## 10 0.9 0.2957746
## 11 1.0 0.2957746## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.OOB"
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 664
## 2 Y 83
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 95
## 2 158
## Reference
## Prediction N Y
## N 664 83
## Y 95 158
## over50k.fctr over50k.fctr.predict.All.X.cp.0.rpart.N
## 1 N 664
## 2 Y 83
## over50k.fctr.predict.All.X.cp.0.rpart.Y
## 1 95
## 2 158
## Prediction
## Reference N Y
## N 664 95
## Y 83 158
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.220000e-01 5.215748e-01 7.968549e-01 8.452324e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 9.132554e-07 4.096641e-01
## model_id model_method
## 1 All.X.cp.0.rpart rpart
## feats
## 1 age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 0 0.719 0.242
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.8936906 0.4 0.7326316 0.873
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.85076 0.8930157 0.6493766 0.8506005
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.3 0.6396761 0.822
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.7968549 0.8452324 0.5215748
## [1] "iterating over method:rf"
## [1] "fitting model: All.X.rf"
## [1] " indep_vars: age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr, .rnorm"## Loading required package: randomForest
## randomForest 4.6-10
## Type rfNews() to see new features/changes/bug fixes.
## + : mtry= 2
## - : mtry= 2
## + : mtry=50
## - : mtry=50
## + : mtry=98
## - : mtry=98
## Aggregating results
## Selecting tuning parameters
## Fitting mtry = 98 on full training set## Warning in myfit_mdl(model_id = paste0(model_id_pfx, ""), model_method =
## method, : model's bestTune found at an extreme of tuneGrid for parameter:
## mtry
## Length Class Mode
## call 4 -none- call
## type 1 -none- character
## predicted 1000 factor numeric
## err.rate 1500 -none- numeric
## confusion 6 -none- numeric
## votes 2000 matrix numeric
## oob.times 1000 -none- numeric
## classes 2 -none- character
## importance 98 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 1000 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 98 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.All.X.rf.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 581 0
## Y 178 241
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 581
## 2 Y 0
## over50k.fctr.predict.All.X.rf.Y
## 1 178
## 2 241
## Reference
## Prediction N Y
## N 706 0
## Y 53 241
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 706
## 2 Y 0
## over50k.fctr.predict.All.X.rf.Y
## 1 53
## 2 241
## Reference
## Prediction N Y
## N 750 0
## Y 9 241
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 750
## 2 Y 0
## over50k.fctr.predict.All.X.rf.Y
## 1 9
## 2 241
## Reference
## Prediction N Y
## N 759 0
## Y 0 241
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 759
## 2 Y 0
## over50k.fctr.predict.All.X.rf.Y
## 1 0
## 2 241
## Reference
## Prediction N Y
## N 759 0
## Y 0 241
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 759
## 2 Y 0
## over50k.fctr.predict.All.X.rf.Y
## 1 0
## 2 241
## Reference
## Prediction N Y
## N 759 0
## Y 0 241
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 759
## 2 Y 0
## over50k.fctr.predict.All.X.rf.Y
## 1 0
## 2 241
## Reference
## Prediction N Y
## N 759 38
## Y 0 203
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 759
## 2 Y 38
## over50k.fctr.predict.All.X.rf.Y
## 1 0
## 2 203
## Reference
## Prediction N Y
## N 759 97
## Y 0 144
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 759
## 2 Y 97
## over50k.fctr.predict.All.X.rf.Y
## 1 0
## 2 144
## Reference
## Prediction N Y
## N 759 155
## Y 0 86
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 759
## 2 Y 155
## over50k.fctr.predict.All.X.rf.Y
## 1 0
## 2 86
## Reference
## Prediction N Y
## N 759 223
## Y 0 18
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 759
## 2 Y 223
## over50k.fctr.predict.All.X.rf.Y
## 1 0
## 2 18
## threshold f.score
## 1 0.0 0.3883965
## 2 0.1 0.7303030
## 3 0.2 0.9009346
## 4 0.3 0.9816701
## 5 0.4 1.0000000
## 6 0.5 1.0000000
## 7 0.6 1.0000000
## 8 0.7 0.9144144
## 9 0.8 0.7480519
## 10 0.9 0.5259939
## 11 1.0 0.1389961
## [1] "Classifier Probability Threshold: 0.6000 to maximize f.score.fit"
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 759
## 2 Y NA
## over50k.fctr.predict.All.X.rf.Y
## 1 NA
## 2 241
## Reference
## Prediction N Y
## N 759 0
## Y 0 241
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 759
## 2 Y 0
## over50k.fctr.predict.All.X.rf.Y
## 1 0
## 2 241
## Prediction
## Reference N Y
## N 759 0
## Y 0 241
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 1.000000e+00 1.000000e+00 9.963179e-01 1.000000e+00 7.590000e-01
## AccuracyPValue McnemarPValue
## 1.744922e-120 NaN
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.All.X.rf.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 483 19
## Y 276 222
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 483
## 2 Y 19
## over50k.fctr.predict.All.X.rf.Y
## 1 276
## 2 222
## Reference
## Prediction N Y
## N 555 37
## Y 204 204
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 555
## 2 Y 37
## over50k.fctr.predict.All.X.rf.Y
## 1 204
## 2 204
## Reference
## Prediction N Y
## N 608 58
## Y 151 183
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 608
## 2 Y 58
## over50k.fctr.predict.All.X.rf.Y
## 1 151
## 2 183
## Reference
## Prediction N Y
## N 658 77
## Y 101 164
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 658
## 2 Y 77
## over50k.fctr.predict.All.X.rf.Y
## 1 101
## 2 164
## Reference
## Prediction N Y
## N 696 96
## Y 63 145
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 696
## 2 Y 96
## over50k.fctr.predict.All.X.rf.Y
## 1 63
## 2 145
## Reference
## Prediction N Y
## N 718 119
## Y 41 122
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 718
## 2 Y 119
## over50k.fctr.predict.All.X.rf.Y
## 1 41
## 2 122
## Reference
## Prediction N Y
## N 735 144
## Y 24 97
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 735
## 2 Y 144
## over50k.fctr.predict.All.X.rf.Y
## 1 24
## 2 97
## Reference
## Prediction N Y
## N 743 175
## Y 16 66
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 743
## 2 Y 175
## over50k.fctr.predict.All.X.rf.Y
## 1 16
## 2 66
## Reference
## Prediction N Y
## N 755 194
## Y 4 47
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 755
## 2 Y 194
## over50k.fctr.predict.All.X.rf.Y
## 1 4
## 2 47
## Reference
## Prediction N Y
## N 759 234
## Y 0 7
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 759
## 2 Y 234
## over50k.fctr.predict.All.X.rf.Y
## 1 0
## 2 7
## threshold f.score
## 1 0.0 0.38839645
## 2 0.1 0.60081191
## 3 0.2 0.62865948
## 4 0.3 0.63652174
## 5 0.4 0.64822134
## 6 0.5 0.64587973
## 7 0.6 0.60396040
## 8 0.7 0.53591160
## 9 0.8 0.40866873
## 10 0.9 0.32191781
## 11 1.0 0.05645161
## [1] "Classifier Probability Threshold: 0.4000 to maximize f.score.OOB"
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 658
## 2 Y 77
## over50k.fctr.predict.All.X.rf.Y
## 1 101
## 2 164
## Reference
## Prediction N Y
## N 658 77
## Y 101 164
## over50k.fctr over50k.fctr.predict.All.X.rf.N
## 1 N 658
## 2 Y 77
## over50k.fctr.predict.All.X.rf.Y
## 1 101
## 2 164
## Prediction
## Reference N Y
## N 658 101
## Y 77 164
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.220000e-01 5.294366e-01 7.968549e-01 8.452324e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 9.132554e-07 8.472177e-02
## model_id model_method
## 1 All.X.rf rf
## feats
## 1 age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr, .rnorm
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 3 23.395 6.721
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 1 0.6 1 0.829
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.9963179 1 0.4992562 0.8749638
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.4 0.6482213 0.822
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.7968549 0.8452324 0.5294366
## [1] "fitting model: All.X.no.rnorm.rf"
## [1] " indep_vars: age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr"
## + : mtry= 2
## - : mtry= 2
## + : mtry=49
## - : mtry=49
## + : mtry=97
## - : mtry=97
## Aggregating results
## Selecting tuning parameters
## Fitting mtry = 49 on full training set
## Length Class Mode
## call 4 -none- call
## type 1 -none- character
## predicted 1000 factor numeric
## err.rate 1500 -none- numeric
## confusion 6 -none- numeric
## votes 2000 matrix numeric
## oob.times 1000 -none- numeric
## classes 2 -none- character
## importance 97 -none- numeric
## importanceSD 0 -none- NULL
## localImportance 0 -none- NULL
## proximity 0 -none- NULL
## ntree 1 -none- numeric
## mtry 1 -none- numeric
## forest 14 -none- list
## y 1000 factor numeric
## test 0 -none- NULL
## inbag 0 -none- NULL
## xNames 97 -none- character
## problemType 1 -none- character
## tuneValue 1 data.frame list
## obsLevels 2 -none- character
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 590 0
## Y 169 241
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 590
## 2 Y 0
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 169
## 2 241
## Reference
## Prediction N Y
## N 689 0
## Y 70 241
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 689
## 2 Y 0
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 70
## 2 241
## Reference
## Prediction N Y
## N 737 0
## Y 22 241
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 737
## 2 Y 0
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 22
## 2 241
## Reference
## Prediction N Y
## N 758 1
## Y 1 240
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 758
## 2 Y 1
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 1
## 2 240
## Reference
## Prediction N Y
## N 759 2
## Y 0 239
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 759
## 2 Y 2
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 0
## 2 239
## Reference
## Prediction N Y
## N 759 8
## Y 0 233
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 759
## 2 Y 8
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 0
## 2 233
## Reference
## Prediction N Y
## N 759 45
## Y 0 196
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 759
## 2 Y 45
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 0
## 2 196
## Reference
## Prediction N Y
## N 759 101
## Y 0 140
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 759
## 2 Y 101
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 0
## 2 140
## Reference
## Prediction N Y
## N 759 153
## Y 0 88
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 759
## 2 Y 153
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 0
## 2 88
## Reference
## Prediction N Y
## N 759 240
## Y 0 1
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 759
## 2 Y 240
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 0
## 2 1
## threshold f.score
## 1 0.0 0.388396454
## 2 0.1 0.740399386
## 3 0.2 0.873188406
## 4 0.3 0.956349206
## 5 0.4 0.995850622
## 6 0.5 0.995833333
## 7 0.6 0.983122363
## 8 0.7 0.897025172
## 9 0.8 0.734908136
## 10 0.9 0.534954407
## 11 1.0 0.008264463
## [1] "Classifier Probability Threshold: 0.4000 to maximize f.score.fit"
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 758
## 2 Y 1
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 1
## 2 240
## Reference
## Prediction N Y
## N 758 1
## Y 1 240
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 758
## 2 Y 1
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 1
## 2 240
## Prediction
## Reference N Y
## N 758 1
## Y 1 240
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 9.980000e-01 9.945331e-01 9.927942e-01 9.997577e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 8.843011e-116 1.000000e+00
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 496 22
## Y 263 219
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 496
## 2 Y 22
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 263
## 2 219
## Reference
## Prediction N Y
## N 557 37
## Y 202 204
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 557
## 2 Y 37
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 202
## 2 204
## Reference
## Prediction N Y
## N 620 52
## Y 139 189
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 620
## 2 Y 52
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 139
## 2 189
## Reference
## Prediction N Y
## N 661 77
## Y 98 164
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 661
## 2 Y 77
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 98
## 2 164
## Reference
## Prediction N Y
## N 699 92
## Y 60 149
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 699
## 2 Y 92
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 60
## 2 149
## Reference
## Prediction N Y
## N 717 117
## Y 42 124
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 717
## 2 Y 117
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 42
## 2 124
## Reference
## Prediction N Y
## N 738 141
## Y 21 100
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 738
## 2 Y 141
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 21
## 2 100
## Reference
## Prediction N Y
## N 742 164
## Y 17 77
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 742
## 2 Y 164
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 17
## 2 77
## Reference
## Prediction N Y
## N 753 192
## Y 6 49
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 753
## 2 Y 192
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 6
## 2 49
## Reference
## Prediction N Y
## N 759 239
## Y 0 2
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 759
## 2 Y 239
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 0
## 2 2
## threshold f.score
## 1 0.0 0.38839645
## 2 0.1 0.60580913
## 3 0.2 0.63060278
## 4 0.3 0.66432337
## 5 0.4 0.65208748
## 6 0.5 0.66222222
## 7 0.6 0.60933661
## 8 0.7 0.55248619
## 9 0.8 0.45970149
## 10 0.9 0.33108108
## 11 1.0 0.01646091
## [1] "Classifier Probability Threshold: 0.3000 to maximize f.score.OOB"
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 620
## 2 Y 52
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 139
## 2 189
## Reference
## Prediction N Y
## N 620 52
## Y 139 189
## over50k.fctr over50k.fctr.predict.All.X.no.rnorm.rf.N
## 1 N 620
## 2 Y 52
## over50k.fctr.predict.All.X.no.rnorm.rf.Y
## 1 139
## 2 189
## Prediction
## Reference N Y
## N 620 139
## Y 52 189
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.090000e-01 5.351712e-01 7.832406e-01 8.329252e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 8.884330e-05 4.885416e-10
## model_id model_method
## 1 All.X.no.rnorm.rf rf
## feats
## 1 age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 3 20.399 5.049
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.9999891 0.4 0.9958506 0.828
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.9927942 0.9997577 0.5147302 0.8805182
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.3 0.6643234 0.809
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.7832406 0.8329252 0.5351712# Simplify a model
# fit_df <- glb_trnent_df; glb_mdl <- step(<complex>_mdl)
print(glb_models_df)## model_id model_method
## 1 MFO.myMFO_classfr myMFO_classfr
## 2 Random.myrandom_classfr myrandom_classfr
## 3 Max.cor.Y.cv.0.rpart rpart
## 4 Max.cor.Y.cv.0.cp.0.rpart rpart
## 5 Max.cor.Y.rpart rpart
## 6 Max.cor.Y.glm glm
## 7 Low.cor.X.glm glm
## 8 All.X.glm glm
## 9 All.X.rpart rpart
## 10 All.X.cp.0.rpart rpart
## 11 All.X.rf rf
## 12 All.X.no.rnorm.rf rf
## feats
## 1 .rnorm
## 2 .rnorm
## 3 age
## 4 age
## 5 age
## 6 age
## 7 age, capitalgain, hoursperweek, capitalloss, workclass.fctr, maritalstatus.fctr, race.fctr, .rnorm, education.fctr, nativecountry.fctr, occupation.fctr, relationship.fctr, sex.fctr
## 8 age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr, .rnorm
## 9 age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr
## 10 age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr
## 11 age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr, .rnorm
## 12 age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 0 0.372 0.002
## 2 0 0.223 0.001
## 3 0 0.779 0.020
## 4 0 0.668 0.018
## 5 3 0.978 0.021
## 6 1 1.074 0.019
## 7 1 2.830 0.499
## 8 1 3.119 0.496
## 9 3 2.392 0.247
## 10 0 0.719 0.242
## 11 3 23.395 6.721
## 12 3 20.399 5.049
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.5000000 0.5 0.0000000 0.7590000
## 2 0.5010961 0.2 0.3883965 0.2410000
## 3 0.5000000 0.5 0.0000000 0.7590000
## 4 0.7183097 0.2 0.4791878 0.4870000
## 5 0.5000000 0.5 0.0000000 0.7459975
## 6 0.6880368 0.2 0.4728192 0.7550095
## 7 0.9325658 0.4 0.7360000 0.8299947
## 8 0.9325658 0.4 0.7360000 0.8299947
## 9 0.8336586 0.3 0.6172466 0.8020056
## 10 0.8936906 0.4 0.7326316 0.8730000
## 11 1.0000000 0.6 1.0000000 0.8290000
## 12 0.9999891 0.4 0.9958506 0.8280000
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit max.auc.OOB
## 1 0.7312521 0.7852137 0.00000000 0.5000000
## 2 0.2147863 0.2687479 0.00000000 0.4879209
## 3 0.7312521 0.7852137 0.00000000 0.5000000
## 4 0.4555980 0.5184787 0.18102402 0.6837507
## 5 0.7312521 0.7852137 0.05249700 0.5000000
## 6 0.5517330 0.6137769 0.02934095 0.6937196
## 7 0.8454346 0.8883720 0.52918164 0.8353370
## 8 0.8454346 0.8883720 0.52918164 0.8353370
## 9 0.7188527 0.7736848 0.34051565 0.8063651
## 10 0.8507600 0.8930157 0.64937661 0.8506005
## 11 0.9963179 1.0000000 0.49925621 0.8749638
## 12 0.9927942 0.9997577 0.51473020 0.8805182
## opt.prob.threshold.OOB max.f.score.OOB max.Accuracy.OOB
## 1 0.5 0.0000000 0.759
## 2 0.2 0.3883965 0.241
## 3 0.5 0.0000000 0.759
## 4 0.2 0.4763903 0.501
## 5 0.5 0.0000000 0.759
## 6 0.2 0.4916773 0.603
## 7 0.4 0.6404959 0.826
## 8 0.4 0.6404959 0.826
## 9 0.3 0.5987842 0.736
## 10 0.3 0.6396761 0.822
## 11 0.4 0.6482213 0.822
## 12 0.3 0.6643234 0.809
## max.AccuracyLower.OOB max.AccuracyUpper.OOB max.Kappa.OOB
## 1 0.7312521 0.7852137 0.0000000
## 2 0.2147863 0.2687479 0.0000000
## 3 0.7312521 0.7852137 0.0000000
## 4 0.4695463 0.5324478 0.1817204
## 5 0.7312521 0.7852137 0.0000000
## 6 0.5719080 0.6334837 0.2375941
## 7 0.8010531 0.8490099 0.5257227
## 8 0.8010531 0.8490099 0.5257227
## 9 0.7075104 0.7630926 0.4223271
## 10 0.7968549 0.8452324 0.5215748
## 11 0.7968549 0.8452324 0.5294366
## 12 0.7832406 0.8329252 0.5351712
## max.AccuracySD.fit max.KappaSD.fit min.aic.fit
## 1 NA NA NA
## 2 NA NA NA
## 3 NA NA NA
## 4 NA NA NA
## 5 0.012206415 0.02765327 NA
## 6 0.008758591 0.03529993 1048.0810
## 7 0.052755235 0.12923177 719.1617
## 8 0.052755235 0.12923177 719.1617
## 9 0.005703043 0.10046165 NA
## 10 NA NA NA
## 11 NA NA NA
## 12 NA NA NAglb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="fit.models",
chunk_step_major=glb_script_df[nrow(glb_script_df), "chunk_step_major"],
chunk_step_minor=glb_script_df[nrow(glb_script_df), "chunk_step_minor"]+1,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed8 fit.models 5 0 7.989
## elapsed9 fit.models 5 1 98.841if (!is.null(glb_model_metric_smmry)) {
stats_df <- glb_models_df[, "model_id", FALSE]
stats_mdl_df <- data.frame()
for (model_id in stats_df$model_id) {
stats_mdl_df <- rbind(stats_mdl_df,
mypredict_mdl(glb_models_lst[[model_id]], glb_trnent_df, glb_rsp_var,
glb_rsp_var_out, model_id, "fit",
glb_model_metric_smmry, glb_model_metric,
glb_model_metric_maximize, ret_type="stats"))
}
stats_df <- merge(stats_df, stats_mdl_df, all.x=TRUE)
stats_mdl_df <- data.frame()
for (model_id in stats_df$model_id) {
stats_mdl_df <- rbind(stats_mdl_df,
mypredict_mdl(glb_models_lst[[model_id]], glb_newent_df, glb_rsp_var,
glb_rsp_var_out, model_id, "OOB",
glb_model_metric_smmry, glb_model_metric,
glb_model_metric_maximize, ret_type="stats"))
}
stats_df <- merge(stats_df, stats_mdl_df, all.x=TRUE)
# tmp_models_df <- orderBy(~model_id, glb_models_df)
# rownames(tmp_models_df) <- seq(1, nrow(tmp_models_df))
# all.equal(subset(tmp_models_df[, names(stats_df)], model_id != "Random.myrandom_classfr"),
# subset(stats_df, model_id != "Random.myrandom_classfr"))
# print(subset(tmp_models_df[, names(stats_df)], model_id != "Random.myrandom_classfr")[, c("model_id", "max.Accuracy.fit")])
# print(subset(stats_df, model_id != "Random.myrandom_classfr")[, c("model_id", "max.Accuracy.fit")])
print("Merging following data into glb_models_df:")
print(stats_mrg_df <- stats_df[, c(1, grep(glb_model_metric, names(stats_df)))])
print(tmp_models_df <- orderBy(~model_id, glb_models_df[, c("model_id", grep(glb_model_metric, names(stats_df), value=TRUE))]))
tmp2_models_df <- glb_models_df[, c("model_id", setdiff(names(glb_models_df), grep(glb_model_metric, names(stats_df), value=TRUE)))]
tmp3_models_df <- merge(tmp2_models_df, stats_mrg_df, all.x=TRUE, sort=FALSE)
print(tmp3_models_df)
print(names(tmp3_models_df))
print(glb_models_df <- subset(tmp3_models_df, select=-model_id.1))
}
plt_models_df <- glb_models_df[, -grep("SD|Upper|Lower", names(glb_models_df))]
for (var in grep("^min.", names(plt_models_df), value=TRUE)) {
plt_models_df[, sub("min.", "inv.", var)] <-
#ifelse(all(is.na(tmp <- plt_models_df[, var])), NA, 1.0 / tmp)
1.0 / plt_models_df[, var]
plt_models_df <- plt_models_df[ , -grep(var, names(plt_models_df))]
}
print(plt_models_df)## model_id model_method
## 1 MFO.myMFO_classfr myMFO_classfr
## 2 Random.myrandom_classfr myrandom_classfr
## 3 Max.cor.Y.cv.0.rpart rpart
## 4 Max.cor.Y.cv.0.cp.0.rpart rpart
## 5 Max.cor.Y.rpart rpart
## 6 Max.cor.Y.glm glm
## 7 Low.cor.X.glm glm
## 8 All.X.glm glm
## 9 All.X.rpart rpart
## 10 All.X.cp.0.rpart rpart
## 11 All.X.rf rf
## 12 All.X.no.rnorm.rf rf
## feats
## 1 .rnorm
## 2 .rnorm
## 3 age
## 4 age
## 5 age
## 6 age
## 7 age, capitalgain, hoursperweek, capitalloss, workclass.fctr, maritalstatus.fctr, race.fctr, .rnorm, education.fctr, nativecountry.fctr, occupation.fctr, relationship.fctr, sex.fctr
## 8 age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr, .rnorm
## 9 age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr
## 10 age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr
## 11 age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr, .rnorm
## 12 age, capitalgain, capitalloss, hoursperweek, workclass.fctr, education.fctr, maritalstatus.fctr, occupation.fctr, relationship.fctr, race.fctr, sex.fctr, nativecountry.fctr
## max.nTuningRuns max.auc.fit opt.prob.threshold.fit max.f.score.fit
## 1 0 0.5000000 0.5 0.0000000
## 2 0 0.5010961 0.2 0.3883965
## 3 0 0.5000000 0.5 0.0000000
## 4 0 0.7183097 0.2 0.4791878
## 5 3 0.5000000 0.5 0.0000000
## 6 1 0.6880368 0.2 0.4728192
## 7 1 0.9325658 0.4 0.7360000
## 8 1 0.9325658 0.4 0.7360000
## 9 3 0.8336586 0.3 0.6172466
## 10 0 0.8936906 0.4 0.7326316
## 11 3 1.0000000 0.6 1.0000000
## 12 3 0.9999891 0.4 0.9958506
## max.Accuracy.fit max.Kappa.fit max.auc.OOB opt.prob.threshold.OOB
## 1 0.7590000 0.00000000 0.5000000 0.5
## 2 0.2410000 0.00000000 0.4879209 0.2
## 3 0.7590000 0.00000000 0.5000000 0.5
## 4 0.4870000 0.18102402 0.6837507 0.2
## 5 0.7459975 0.05249700 0.5000000 0.5
## 6 0.7550095 0.02934095 0.6937196 0.2
## 7 0.8299947 0.52918164 0.8353370 0.4
## 8 0.8299947 0.52918164 0.8353370 0.4
## 9 0.8020056 0.34051565 0.8063651 0.3
## 10 0.8730000 0.64937661 0.8506005 0.3
## 11 0.8290000 0.49925621 0.8749638 0.4
## 12 0.8280000 0.51473020 0.8805182 0.3
## max.f.score.OOB max.Accuracy.OOB max.Kappa.OOB
## 1 0.0000000 0.759 0.0000000
## 2 0.3883965 0.241 0.0000000
## 3 0.0000000 0.759 0.0000000
## 4 0.4763903 0.501 0.1817204
## 5 0.0000000 0.759 0.0000000
## 6 0.4916773 0.603 0.2375941
## 7 0.6404959 0.826 0.5257227
## 8 0.6404959 0.826 0.5257227
## 9 0.5987842 0.736 0.4223271
## 10 0.6396761 0.822 0.5215748
## 11 0.6482213 0.822 0.5294366
## 12 0.6643234 0.809 0.5351712
## inv.elapsedtime.everything inv.elapsedtime.final inv.aic.fit
## 1 2.68817204 500.0000000 NA
## 2 4.48430493 1000.0000000 NA
## 3 1.28369705 50.0000000 NA
## 4 1.49700599 55.5555556 NA
## 5 1.02249489 47.6190476 NA
## 6 0.93109870 52.6315789 0.0009541247
## 7 0.35335689 2.0040080 0.0013905079
## 8 0.32061558 2.0161290 0.0013905079
## 9 0.41806020 4.0485830 NA
## 10 1.39082058 4.1322314 NA
## 11 0.04274418 0.1487874 NA
## 12 0.04902201 0.1980590 NAprint(myplot_radar(radar_inp_df=plt_models_df))## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors## Warning: The shape palette can deal with a maximum of 6 discrete values
## because more than 6 becomes difficult to discriminate; you have
## 12. Consider specifying shapes manually. if you must have them.## Warning: Removed 6 rows containing missing values (geom_path).## Warning: Removed 89 rows containing missing values (geom_point).## Warning: Removed 9 rows containing missing values (geom_text).## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors## Warning: The shape palette can deal with a maximum of 6 discrete values
## because more than 6 becomes difficult to discriminate; you have
## 12. Consider specifying shapes manually. if you must have them.
# print(myplot_radar(radar_inp_df=subset(plt_models_df,
# !(model_id %in% grep("random|MFO", plt_models_df$model_id, value=TRUE)))))
# Compute CI for <metric>SD
glb_models_df <- mutate(glb_models_df,
max.df = ifelse(max.nTuningRuns > 1, max.nTuningRuns - 1, NA),
min.sd2ci.scaler = ifelse(is.na(max.df), NA, qt(0.975, max.df)))
for (var in grep("SD", names(glb_models_df), value=TRUE)) {
# Does CI alredy exist ?
var_components <- unlist(strsplit(var, "SD"))
varActul <- paste0(var_components[1], var_components[2])
varUpper <- paste0(var_components[1], "Upper", var_components[2])
varLower <- paste0(var_components[1], "Lower", var_components[2])
if (varUpper %in% names(glb_models_df)) {
warning(varUpper, " already exists in glb_models_df")
# Assuming Lower also exists
next
}
print(sprintf("var:%s", var))
# CI is dependent on sample size in t distribution; df=n-1
glb_models_df[, varUpper] <- glb_models_df[, varActul] +
glb_models_df[, "min.sd2ci.scaler"] * glb_models_df[, var]
glb_models_df[, varLower] <- glb_models_df[, varActul] -
glb_models_df[, "min.sd2ci.scaler"] * glb_models_df[, var]
}## Warning: max.AccuracyUpper.fit already exists in glb_models_df## [1] "var:max.KappaSD.fit"# Plot metrics with CI
plt_models_df <- glb_models_df[, "model_id", FALSE]
pltCI_models_df <- glb_models_df[, "model_id", FALSE]
for (var in grep("Upper", names(glb_models_df), value=TRUE)) {
var_components <- unlist(strsplit(var, "Upper"))
col_name <- unlist(paste(var_components, collapse=""))
plt_models_df[, col_name] <- glb_models_df[, col_name]
for (name in paste0(var_components[1], c("Upper", "Lower"), var_components[2]))
pltCI_models_df[, name] <- glb_models_df[, name]
}
build_statsCI_data <- function(plt_models_df) {
mltd_models_df <- melt(plt_models_df, id.vars="model_id")
mltd_models_df$data <- sapply(1:nrow(mltd_models_df),
function(row_ix) tail(unlist(strsplit(as.character(
mltd_models_df[row_ix, "variable"]), "[.]")), 1))
mltd_models_df$label <- sapply(1:nrow(mltd_models_df),
function(row_ix) head(unlist(strsplit(as.character(
mltd_models_df[row_ix, "variable"]), paste0(".", mltd_models_df[row_ix, "data"]))), 1))
#print(mltd_models_df)
return(mltd_models_df)
}
mltd_models_df <- build_statsCI_data(plt_models_df)
mltdCI_models_df <- melt(pltCI_models_df, id.vars="model_id")
for (row_ix in 1:nrow(mltdCI_models_df)) {
for (type in c("Upper", "Lower")) {
if (length(var_components <- unlist(strsplit(
as.character(mltdCI_models_df[row_ix, "variable"]), type))) > 1) {
#print(sprintf("row_ix:%d; type:%s; ", row_ix, type))
mltdCI_models_df[row_ix, "label"] <- var_components[1]
mltdCI_models_df[row_ix, "data"] <- unlist(strsplit(var_components[2], "[.]"))[2]
mltdCI_models_df[row_ix, "type"] <- type
break
}
}
}
#print(mltdCI_models_df)
# castCI_models_df <- dcast(mltdCI_models_df, value ~ type, fun.aggregate=sum)
# print(castCI_models_df)
wideCI_models_df <- reshape(subset(mltdCI_models_df, select=-variable),
timevar="type",
idvar=setdiff(names(mltdCI_models_df), c("type", "value", "variable")),
direction="wide")
#print(wideCI_models_df)
mrgdCI_models_df <- merge(wideCI_models_df, mltd_models_df, all.x=TRUE)
#print(mrgdCI_models_df)
# Merge stats back in if CIs don't exist
goback_vars <- c()
for (var in unique(mltd_models_df$label)) {
for (type in unique(mltd_models_df$data)) {
var_type <- paste0(var, ".", type)
# if this data is already present, next
if (var_type %in% unique(paste(mltd_models_df$label, mltd_models_df$data, sep=".")))
next
#print(sprintf("var_type:%s", var_type))
goback_vars <- c(goback_vars, var_type)
}
}
if (length(goback_vars) > 0) {
mltd_goback_df <- build_statsCI_data(glb_models_df[, c("model_id", goback_vars)])
mltd_models_df <- rbind(mltd_models_df, mltd_goback_df)
}
mltd_models_df <- merge(mltd_models_df, glb_models_df[, c("model_id", "model_method")], all.x=TRUE)
# print(myplot_bar(mltd_models_df, "model_id", "value", colorcol_name="data") +
# geom_errorbar(data=mrgdCI_models_df,
# mapping=aes(x=model_id, ymax=value.Upper, ymin=value.Lower), width=0.5) +
# facet_grid(label ~ data, scales="free") +
# theme(axis.text.x = element_text(angle = 45,vjust = 1)))
# mltd_models_df <- orderBy(~ value +variable +data +label + model_method + model_id,
# mltd_models_df)
print(myplot_bar(mltd_models_df, "model_id", "value", colorcol_name="model_method") +
geom_errorbar(data=mrgdCI_models_df,
mapping=aes(x=model_id, ymax=value.Upper, ymin=value.Lower), width=0.5) +
facet_grid(label ~ data, scales="free") +
theme(axis.text.x = element_text(angle = 90,vjust = 0.5)))
model_evl_terms <- c(NULL)
for (metric in glb_model_evl_criteria)
model_evl_terms <- c(model_evl_terms,
ifelse(length(grep("max", metric)) > 0, "-", "+"), metric)
model_sel_frmla <- as.formula(paste(c("~ ", model_evl_terms), collapse=" "))
print(tmp_models_df <- orderBy(model_sel_frmla, glb_models_df)[, c("model_id", glb_model_evl_criteria)])## model_id max.Accuracy.OOB max.Kappa.OOB min.aic.fit
## 7 Low.cor.X.glm 0.826 0.5257227 719.1617
## 8 All.X.glm 0.826 0.5257227 719.1617
## 11 All.X.rf 0.822 0.5294366 NA
## 10 All.X.cp.0.rpart 0.822 0.5215748 NA
## 12 All.X.no.rnorm.rf 0.809 0.5351712 NA
## 1 MFO.myMFO_classfr 0.759 0.0000000 NA
## 3 Max.cor.Y.cv.0.rpart 0.759 0.0000000 NA
## 5 Max.cor.Y.rpart 0.759 0.0000000 NA
## 9 All.X.rpart 0.736 0.4223271 NA
## 6 Max.cor.Y.glm 0.603 0.2375941 1048.0810
## 4 Max.cor.Y.cv.0.cp.0.rpart 0.501 0.1817204 NA
## 2 Random.myrandom_classfr 0.241 0.0000000 NAprint(myplot_radar(radar_inp_df=tmp_models_df))## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors## Warning: The shape palette can deal with a maximum of 6 discrete values
## because more than 6 becomes difficult to discriminate; you have
## 12. Consider specifying shapes manually. if you must have them.## Warning: Removed 6 rows containing missing values (geom_path).## Warning: Removed 23 rows containing missing values (geom_point).## Warning: Removed 9 rows containing missing values (geom_text).## Warning in RColorBrewer::brewer.pal(n, pal): n too large, allowed maximum for palette Set1 is 9
## Returning the palette you asked for with that many colors## Warning: The shape palette can deal with a maximum of 6 discrete values
## because more than 6 becomes difficult to discriminate; you have
## 12. Consider specifying shapes manually. if you must have them.
print("Metrics used for model selection:"); print(model_sel_frmla)## [1] "Metrics used for model selection:"## ~-max.Accuracy.OOB - max.Kappa.OOB + min.aic.fitprint(sprintf("Best model id: %s", tmp_models_df[1, "model_id"]))## [1] "Best model id: Low.cor.X.glm"if (is.null(glb_sel_mdl_id))
{ glb_sel_mdl_id <- tmp_models_df[1, "model_id"] } else
print(sprintf("User specified selection: %s", glb_sel_mdl_id))
myprint_mdl(glb_sel_mdl <- glb_models_lst[[glb_sel_mdl_id]])## Warning: not plotting observations with leverage one:
## 118, 160, 182, 194, 371, 422, 477, 518, 535, 638, 739, 798, 805, 808, 811, 921
## Warning: not plotting observations with leverage one:
## 118, 160, 182, 194, 371, 422, 477, 518, 535, 638, 739, 798, 805, 808, 811, 921
## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced
##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.24059 -0.43840 -0.08362 0.00000 2.62797
##
## Coefficients: (12 not defined because of singularities)
## Estimate Std. Error
## (Intercept) -6.821e+00 1.159e+00
## age 3.818e-02 1.089e-02
## capitalgain 4.161e-04 7.105e-05
## hoursperweek 2.970e-02 1.051e-02
## capitalloss 2.194e-04 2.357e-04
## `workclass.fctr Self-emp-not-inc` 1.610e-01 7.478e-01
## `workclass.fctr Private` 1.125e+00 6.789e-01
## `workclass.fctr Federal-gov` 2.904e+00 9.072e-01
## `workclass.fctr Local-gov` 3.428e-01 7.691e-01
## `workclass.fctr ?` 6.723e-01 9.573e-01
## `workclass.fctr Self-emp-inc` 1.042e+00 8.233e-01
## `workclass.fctr Without-pay` -1.725e+01 1.075e+04
## `workclass.fctr Never-worked` NA NA
## `maritalstatus.fctr Married-civ-spouse` -1.878e+00 7.345e+03
## `maritalstatus.fctr Divorced` 4.708e-01 5.824e-01
## `maritalstatus.fctr Married-spouse-absent` -1.558e+01 2.722e+03
## `maritalstatus.fctr Separated` 4.013e-01 8.565e-01
## `maritalstatus.fctr Married-AF-spouse` -1.751e+01 1.302e+04
## `maritalstatus.fctr Widowed` 1.088e+00 8.744e-01
## `race.fctr Black` 4.090e-01 4.460e-01
## `race.fctr Asian-Pac-Islander` 1.380e+00 1.491e+00
## `race.fctr Amer-Indian-Eskimo` 5.347e-01 1.206e+00
## `race.fctr Other` 1.267e+00 1.487e+00
## .rnorm -3.095e-02 1.171e-01
## `education.fctr HS-grad` -1.559e+00 3.436e-01
## `education.fctr 11th` -1.815e+01 2.222e+03
## `education.fctr Masters` -2.940e-02 4.506e-01
## `education.fctr 9th` -1.666e+00 1.406e+00
## `education.fctr Some-college` -8.444e-01 3.563e-01
## `education.fctr Assoc-acdm` -6.287e-01 7.288e-01
## `education.fctr 7th-8th` -2.846e+00 1.162e+00
## `education.fctr Doctorate` 1.534e+00 9.430e-01
## `education.fctr Assoc-voc` -8.237e-01 5.183e-01
## `education.fctr Prof-school` 1.781e+00 9.115e-01
## `education.fctr 5th-6th` -2.305e+00 1.545e+00
## `education.fctr 10th` -1.951e+00 7.429e-01
## `education.fctr 1st-4th` -2.408e+00 5.929e+03
## `education.fctr Preschool` -2.289e+00 1.097e+04
## `education.fctr 12th` -1.821e+01 2.665e+03
## `nativecountry.fctr Cuba` 2.854e+00 1.496e+00
## `nativecountry.fctr Jamaica` 1.769e-01 1.363e+00
## `nativecountry.fctr India` -1.501e+00 1.781e+00
## `nativecountry.fctr Mexico` -1.704e+01 2.148e+03
## `nativecountry.fctr South` 3.292e-01 1.099e+04
## `nativecountry.fctr Puerto-Rico` -1.631e+01 7.046e+03
## `nativecountry.fctr Honduras` NA NA
## `nativecountry.fctr England` 9.965e-01 1.034e+00
## `nativecountry.fctr Canada` 2.268e+00 2.148e+00
## `nativecountry.fctr Germany` -1.790e+01 4.090e+03
## `nativecountry.fctr Iran` NA NA
## `nativecountry.fctr Philippines` -1.854e+00 2.302e+00
## `nativecountry.fctr Italy` -1.582e+01 1.075e+04
## `nativecountry.fctr Poland` -1.698e+01 1.075e+04
## `nativecountry.fctr Columbia` -1.694e+01 4.154e+03
## `nativecountry.fctr Cambodia` NA NA
## `nativecountry.fctr Thailand` NA NA
## `nativecountry.fctr Ecuador` -1.120e+01 1.075e+04
## `nativecountry.fctr Laos` 5.683e-01 2.207e+00
## `nativecountry.fctr Taiwan` 1.847e+01 7.390e+03
## `nativecountry.fctr Haiti` -1.274e+01 1.075e+04
## `nativecountry.fctr Portugal` -1.658e+01 1.075e+04
## `nativecountry.fctr Dominican-Republic` NA NA
## `nativecountry.fctr El-Salvador` -1.699e+01 3.705e+03
## `nativecountry.fctr France` -2.093e+01 6.381e+03
## `nativecountry.fctr Guatemala` 1.553e+01 1.175e+04
## `nativecountry.fctr China` -2.950e+00 2.128e+00
## `nativecountry.fctr Japan` -1.679e+01 1.075e+04
## `nativecountry.fctr Yugoslavia` NA NA
## `nativecountry.fctr Peru` -1.283e+01 1.075e+04
## `nativecountry.fctr Outlying-US(Guam-USVI-etc)` NA NA
## `nativecountry.fctr Scotland` -1.651e+01 1.075e+04
## `nativecountry.fctr Trinadad&Tobago` -1.849e+01 1.075e+04
## `nativecountry.fctr Greece` -1.473e+01 1.075e+04
## `nativecountry.fctr Nicaragua` -1.745e+01 1.075e+04
## `nativecountry.fctr Vietnam` -1.078e+00 7.665e+03
## `nativecountry.fctr Hong` -1.907e+01 1.075e+04
## `nativecountry.fctr Ireland` -1.925e+01 1.075e+04
## `nativecountry.fctr Hungary` NA NA
## `nativecountry.fctr Holand-Netherlands` NA NA
## `occupation.fctr Exec-managerial` 1.304e+00 4.950e-01
## `occupation.fctr Handlers-cleaners` 2.694e-02 8.208e-01
## `occupation.fctr Prof-specialty` 1.153e+00 5.194e-01
## `occupation.fctr Other-service` -2.126e+00 1.190e+00
## `occupation.fctr Sales` 1.069e+00 5.227e-01
## `occupation.fctr Transport-moving` 1.505e-01 6.237e-01
## `occupation.fctr Farming-fishing` -1.718e+01 1.838e+03
## `occupation.fctr Machine-op-inspct` 3.740e-01 6.315e-01
## `occupation.fctr Tech-support` 1.914e+00 7.198e-01
## `occupation.fctr Craft-repair` 4.127e-01 5.182e-01
## `occupation.fctr ?` NA NA
## `occupation.fctr Protective-serv` -1.078e+00 1.275e+00
## `occupation.fctr Armed-Forces` NA NA
## `occupation.fctr Priv-house-serv` -1.410e+01 4.167e+03
## `relationship.fctr Husband` 4.541e+00 7.345e+03
## `relationship.fctr Wife` 6.133e+00 7.345e+03
## `relationship.fctr Own-child` -9.141e-01 1.018e+00
## `relationship.fctr Unmarried` 1.153e+00 5.278e-01
## `relationship.fctr Other-relative` -1.367e+01 2.263e+03
## `sex.fctr Female` -1.083e+00 4.967e-01
## z value Pr(>|z|)
## (Intercept) -5.886 3.95e-09 ***
## age 3.507 0.000454 ***
## capitalgain 5.857 4.71e-09 ***
## hoursperweek 2.828 0.004691 **
## capitalloss 0.931 0.351975
## `workclass.fctr Self-emp-not-inc` 0.215 0.829567
## `workclass.fctr Private` 1.657 0.097516 .
## `workclass.fctr Federal-gov` 3.201 0.001370 **
## `workclass.fctr Local-gov` 0.446 0.655834
## `workclass.fctr ?` 0.702 0.482461
## `workclass.fctr Self-emp-inc` 1.265 0.205836
## `workclass.fctr Without-pay` -0.002 0.998720
## `workclass.fctr Never-worked` NA NA
## `maritalstatus.fctr Married-civ-spouse` 0.000 0.999796
## `maritalstatus.fctr Divorced` 0.808 0.418880
## `maritalstatus.fctr Married-spouse-absent` -0.006 0.995434
## `maritalstatus.fctr Separated` 0.469 0.639391
## `maritalstatus.fctr Married-AF-spouse` -0.001 0.998927
## `maritalstatus.fctr Widowed` 1.244 0.213532
## `race.fctr Black` 0.917 0.359078
## `race.fctr Asian-Pac-Islander` 0.925 0.354892
## `race.fctr Amer-Indian-Eskimo` 0.443 0.657450
## `race.fctr Other` 0.852 0.394420
## .rnorm -0.264 0.791562
## `education.fctr HS-grad` -4.537 5.71e-06 ***
## `education.fctr 11th` -0.008 0.993482
## `education.fctr Masters` -0.065 0.947977
## `education.fctr 9th` -1.185 0.236034
## `education.fctr Some-college` -2.370 0.017790 *
## `education.fctr Assoc-acdm` -0.863 0.388343
## `education.fctr 7th-8th` -2.450 0.014284 *
## `education.fctr Doctorate` 1.627 0.103707
## `education.fctr Assoc-voc` -1.589 0.112020
## `education.fctr Prof-school` 1.954 0.050691 .
## `education.fctr 5th-6th` -1.492 0.135688
## `education.fctr 10th` -2.626 0.008635 **
## `education.fctr 1st-4th` 0.000 0.999676
## `education.fctr Preschool` 0.000 0.999833
## `education.fctr 12th` -0.007 0.994550
## `nativecountry.fctr Cuba` 1.908 0.056379 .
## `nativecountry.fctr Jamaica` 0.130 0.896757
## `nativecountry.fctr India` -0.843 0.399391
## `nativecountry.fctr Mexico` -0.008 0.993670
## `nativecountry.fctr South` 0.000 0.999976
## `nativecountry.fctr Puerto-Rico` -0.002 0.998153
## `nativecountry.fctr Honduras` NA NA
## `nativecountry.fctr England` 0.964 0.334943
## `nativecountry.fctr Canada` 1.056 0.291095
## `nativecountry.fctr Germany` -0.004 0.996508
## `nativecountry.fctr Iran` NA NA
## `nativecountry.fctr Philippines` -0.805 0.420791
## `nativecountry.fctr Italy` -0.001 0.998826
## `nativecountry.fctr Poland` -0.002 0.998740
## `nativecountry.fctr Columbia` -0.004 0.996746
## `nativecountry.fctr Cambodia` NA NA
## `nativecountry.fctr Thailand` NA NA
## `nativecountry.fctr Ecuador` -0.001 0.999169
## `nativecountry.fctr Laos` 0.258 0.796765
## `nativecountry.fctr Taiwan` 0.002 0.998006
## `nativecountry.fctr Haiti` -0.001 0.999054
## `nativecountry.fctr Portugal` -0.002 0.998770
## `nativecountry.fctr Dominican-Republic` NA NA
## `nativecountry.fctr El-Salvador` -0.005 0.996341
## `nativecountry.fctr France` -0.003 0.997383
## `nativecountry.fctr Guatemala` 0.001 0.998945
## `nativecountry.fctr China` -1.386 0.165625
## `nativecountry.fctr Japan` -0.002 0.998754
## `nativecountry.fctr Yugoslavia` NA NA
## `nativecountry.fctr Peru` -0.001 0.999048
## `nativecountry.fctr Outlying-US(Guam-USVI-etc)` NA NA
## `nativecountry.fctr Scotland` -0.002 0.998775
## `nativecountry.fctr Trinadad&Tobago` -0.002 0.998628
## `nativecountry.fctr Greece` -0.001 0.998907
## `nativecountry.fctr Nicaragua` -0.002 0.998705
## `nativecountry.fctr Vietnam` 0.000 0.999888
## `nativecountry.fctr Hong` -0.002 0.998585
## `nativecountry.fctr Ireland` -0.002 0.998572
## `nativecountry.fctr Hungary` NA NA
## `nativecountry.fctr Holand-Netherlands` NA NA
## `occupation.fctr Exec-managerial` 2.634 0.008448 **
## `occupation.fctr Handlers-cleaners` 0.033 0.973816
## `occupation.fctr Prof-specialty` 2.220 0.026412 *
## `occupation.fctr Other-service` -1.786 0.074026 .
## `occupation.fctr Sales` 2.045 0.040865 *
## `occupation.fctr Transport-moving` 0.241 0.809342
## `occupation.fctr Farming-fishing` -0.009 0.992542
## `occupation.fctr Machine-op-inspct` 0.592 0.553673
## `occupation.fctr Tech-support` 2.658 0.007851 **
## `occupation.fctr Craft-repair` 0.796 0.425807
## `occupation.fctr ?` NA NA
## `occupation.fctr Protective-serv` -0.846 0.397550
## `occupation.fctr Armed-Forces` NA NA
## `occupation.fctr Priv-house-serv` -0.003 0.997300
## `relationship.fctr Husband` 0.001 0.999507
## `relationship.fctr Wife` 0.001 0.999334
## `relationship.fctr Own-child` -0.898 0.369177
## `relationship.fctr Unmarried` 2.183 0.029005 *
## `relationship.fctr Other-relative` -0.006 0.995180
## `sex.fctr Female` -2.180 0.029227 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1104.46 on 999 degrees of freedom
## Residual deviance: 545.16 on 913 degrees of freedom
## AIC: 719.16
##
## Number of Fisher Scoring iterations: 18## [1] TRUEreplay.petrisim(pn=glb_analytics_pn,
replay.trans=(glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
"model.selected")), flip_coord=TRUE)## time trans "bgn " "fit.data.training.all " "predict.data.new " "end "
## 0.0000 multiple enabled transitions: data.training.all data.new model.selected firing: data.training.all
## 1.0000 1 2 1 0 0
## 1.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction firing: data.new
## 2.0000 2 1 1 1 0
## 2.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction data.new.prediction firing: model.selected
## 3.0000 3 0 2 1 0
glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="fit.data.training.all",
chunk_step_major=max(glb_script_df$chunk_step_major)+1,
chunk_step_minor=0,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed9 fit.models 5 1 98.841
## elapsed10 fit.data.training.all 6 0 110.406Step 6: fit.data.training.all
if (!is.null(glb_fin_mdl_id) && (glb_fin_mdl_id %in% names(glb_models_lst))) {
warning("Final model same as user selected model")
glb_fin_mdl <- glb_sel_mdl
} else {
print(mdl_feats_df <- myextract_mdl_feats(sel_mdl=glb_sel_mdl, entity_df=glb_trnent_df))
if ((model_method <- glb_sel_mdl$method) == "custom")
# get actual method from the model_id
model_method <- tail(unlist(strsplit(glb_sel_mdl_id, "[.]")), 1)
# Sync with parameters in mydsutils.R
ret_lst <- myfit_mdl(model_id="Final", model_method=model_method,
indep_vars_vctr=mdl_feats_df$id, model_type=glb_model_type,
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
fit_df=glb_trnent_df, OOB_df=NULL,
# Automate from here
# Issues if glb_sel_mdl$method == "rf" b/c trainControl is "oob"; not "cv"
n_cv_folds=glb_n_cv_folds, tune_models_df=NULL,
model_loss_mtrx=glb_model_metric_terms,
model_summaryFunction=glb_sel_mdl$control$summaryFunction,
model_metric=glb_sel_mdl$metric,
model_metric_maximize=glb_sel_mdl$maximize)
glb_fin_mdl <- glb_models_lst[[length(glb_models_lst)]]
glb_fin_mdl_id <- glb_models_df[length(glb_models_lst), "model_id"]
}## id importance
## capitalgain capitalgain 100.000000
## education.fctr education.fctr 77.459886
## age age 59.871191
## workclass.fctr race.fctr 54.648029
## hoursperweek hoursperweek 48.274918
## occupation.fctr relationship.fctr 45.387086
## relationship.fctr nativecountry.fctr 37.277684
## sex.fctr maritalstatus.fctr 37.226377
## nativecountry.fctr occupation.fctr 32.576969
## maritalstatus.fctr workclass.fctr 21.237208
## capitalloss capitalloss 15.890756
## race.fctr sex.fctr 15.794747
## .rnorm .rnorm 4.511675
## [1] "fitting model: Final.glm"
## [1] " indep_vars: capitalgain, education.fctr, age, race.fctr, hoursperweek, relationship.fctr, nativecountry.fctr, maritalstatus.fctr, occupation.fctr, workclass.fctr, capitalloss, sex.fctr, .rnorm"
## + Fold1: parameter=none## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## - Fold1: parameter=none
## + Fold2: parameter=none## Warning: glm.fit: algorithm did not converge## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## - Fold2: parameter=none
## + Fold3: parameter=none## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## - Fold3: parameter=none
## Aggregating results
## Fitting final model on full training set## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred## Warning: not plotting observations with leverage one:
## 118, 160, 182, 194, 239, 293, 371, 422, 477, 518, 535, 638, 739, 798, 805, 808, 811, 921
## Warning: not plotting observations with leverage one:
## 118, 160, 182, 194, 239, 293, 371, 422, 477, 518, 535, 638, 739, 798, 805, 808, 811, 921
## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced## Warning in sqrt(crit * p * (1 - hh)/hh): NaNs produced##
## Call:
## NULL
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.24059 -0.43840 -0.08362 0.00000 2.62797
##
## Coefficients: (12 not defined because of singularities)
## Estimate Std. Error
## (Intercept) -6.821e+00 1.159e+00
## capitalgain 4.161e-04 7.105e-05
## `education.fctr HS-grad` -1.559e+00 3.436e-01
## `education.fctr 11th` -1.815e+01 2.222e+03
## `education.fctr Masters` -2.940e-02 4.506e-01
## `education.fctr 9th` -1.666e+00 1.406e+00
## `education.fctr Some-college` -8.444e-01 3.563e-01
## `education.fctr Assoc-acdm` -6.287e-01 7.288e-01
## `education.fctr 7th-8th` -2.846e+00 1.162e+00
## `education.fctr Doctorate` 1.534e+00 9.430e-01
## `education.fctr Assoc-voc` -8.237e-01 5.183e-01
## `education.fctr Prof-school` 1.781e+00 9.115e-01
## `education.fctr 5th-6th` -2.305e+00 1.545e+00
## `education.fctr 10th` -1.951e+00 7.429e-01
## `education.fctr 1st-4th` -2.408e+00 5.929e+03
## `education.fctr Preschool` -2.289e+00 1.097e+04
## `education.fctr 12th` -1.821e+01 2.665e+03
## age 3.818e-02 1.089e-02
## `race.fctr Black` 4.090e-01 4.460e-01
## `race.fctr Asian-Pac-Islander` 1.380e+00 1.491e+00
## `race.fctr Amer-Indian-Eskimo` 5.347e-01 1.206e+00
## `race.fctr Other` 1.267e+00 1.487e+00
## hoursperweek 2.970e-02 1.051e-02
## `relationship.fctr Husband` 4.541e+00 7.345e+03
## `relationship.fctr Wife` 6.133e+00 7.345e+03
## `relationship.fctr Own-child` -9.141e-01 1.018e+00
## `relationship.fctr Unmarried` 1.153e+00 5.278e-01
## `relationship.fctr Other-relative` -1.367e+01 2.263e+03
## `nativecountry.fctr Cuba` 2.854e+00 1.496e+00
## `nativecountry.fctr Jamaica` 1.769e-01 1.363e+00
## `nativecountry.fctr India` -1.501e+00 1.781e+00
## `nativecountry.fctr Mexico` -1.704e+01 2.148e+03
## `nativecountry.fctr South` 3.292e-01 1.099e+04
## `nativecountry.fctr Puerto-Rico` -1.631e+01 7.046e+03
## `nativecountry.fctr Honduras` NA NA
## `nativecountry.fctr England` 9.965e-01 1.034e+00
## `nativecountry.fctr Canada` 2.268e+00 2.148e+00
## `nativecountry.fctr Germany` -1.790e+01 4.090e+03
## `nativecountry.fctr Iran` NA NA
## `nativecountry.fctr Philippines` -1.854e+00 2.302e+00
## `nativecountry.fctr Italy` -1.582e+01 1.075e+04
## `nativecountry.fctr Poland` -1.698e+01 1.075e+04
## `nativecountry.fctr Columbia` -1.694e+01 4.154e+03
## `nativecountry.fctr Cambodia` NA NA
## `nativecountry.fctr Thailand` NA NA
## `nativecountry.fctr Ecuador` -1.120e+01 1.075e+04
## `nativecountry.fctr Laos` 5.683e-01 2.207e+00
## `nativecountry.fctr Taiwan` 1.847e+01 7.390e+03
## `nativecountry.fctr Haiti` -1.274e+01 1.075e+04
## `nativecountry.fctr Portugal` -1.658e+01 1.075e+04
## `nativecountry.fctr Dominican-Republic` NA NA
## `nativecountry.fctr El-Salvador` -1.699e+01 3.705e+03
## `nativecountry.fctr France` -2.093e+01 6.381e+03
## `nativecountry.fctr Guatemala` 1.553e+01 1.175e+04
## `nativecountry.fctr China` -2.950e+00 2.128e+00
## `nativecountry.fctr Japan` -1.679e+01 1.075e+04
## `nativecountry.fctr Yugoslavia` NA NA
## `nativecountry.fctr Peru` -1.283e+01 1.075e+04
## `nativecountry.fctr Outlying-US(Guam-USVI-etc)` NA NA
## `nativecountry.fctr Scotland` -1.651e+01 1.075e+04
## `nativecountry.fctr Trinadad&Tobago` -1.849e+01 1.075e+04
## `nativecountry.fctr Greece` -1.473e+01 1.075e+04
## `nativecountry.fctr Nicaragua` -1.745e+01 1.075e+04
## `nativecountry.fctr Vietnam` -1.078e+00 7.665e+03
## `nativecountry.fctr Hong` -1.907e+01 1.075e+04
## `nativecountry.fctr Ireland` -1.925e+01 1.075e+04
## `nativecountry.fctr Hungary` NA NA
## `nativecountry.fctr Holand-Netherlands` NA NA
## `maritalstatus.fctr Married-civ-spouse` -1.878e+00 7.345e+03
## `maritalstatus.fctr Divorced` 4.708e-01 5.824e-01
## `maritalstatus.fctr Married-spouse-absent` -1.558e+01 2.722e+03
## `maritalstatus.fctr Separated` 4.013e-01 8.565e-01
## `maritalstatus.fctr Married-AF-spouse` -1.751e+01 1.302e+04
## `maritalstatus.fctr Widowed` 1.088e+00 8.744e-01
## `occupation.fctr Exec-managerial` 1.304e+00 4.950e-01
## `occupation.fctr Handlers-cleaners` 2.694e-02 8.208e-01
## `occupation.fctr Prof-specialty` 1.153e+00 5.194e-01
## `occupation.fctr Other-service` -2.126e+00 1.190e+00
## `occupation.fctr Sales` 1.069e+00 5.227e-01
## `occupation.fctr Transport-moving` 1.505e-01 6.237e-01
## `occupation.fctr Farming-fishing` -1.718e+01 1.838e+03
## `occupation.fctr Machine-op-inspct` 3.740e-01 6.315e-01
## `occupation.fctr Tech-support` 1.914e+00 7.198e-01
## `occupation.fctr Craft-repair` 4.127e-01 5.182e-01
## `occupation.fctr ?` 6.723e-01 9.573e-01
## `occupation.fctr Protective-serv` -1.078e+00 1.275e+00
## `occupation.fctr Armed-Forces` NA NA
## `occupation.fctr Priv-house-serv` -1.410e+01 4.167e+03
## `workclass.fctr Self-emp-not-inc` 1.610e-01 7.478e-01
## `workclass.fctr Private` 1.125e+00 6.789e-01
## `workclass.fctr Federal-gov` 2.904e+00 9.072e-01
## `workclass.fctr Local-gov` 3.428e-01 7.691e-01
## `workclass.fctr ?` NA NA
## `workclass.fctr Self-emp-inc` 1.042e+00 8.233e-01
## `workclass.fctr Without-pay` -1.725e+01 1.075e+04
## `workclass.fctr Never-worked` NA NA
## capitalloss 2.194e-04 2.357e-04
## `sex.fctr Female` -1.083e+00 4.967e-01
## .rnorm -3.095e-02 1.171e-01
## z value Pr(>|z|)
## (Intercept) -5.886 3.95e-09 ***
## capitalgain 5.857 4.71e-09 ***
## `education.fctr HS-grad` -4.537 5.71e-06 ***
## `education.fctr 11th` -0.008 0.993482
## `education.fctr Masters` -0.065 0.947977
## `education.fctr 9th` -1.185 0.236034
## `education.fctr Some-college` -2.370 0.017790 *
## `education.fctr Assoc-acdm` -0.863 0.388343
## `education.fctr 7th-8th` -2.450 0.014284 *
## `education.fctr Doctorate` 1.627 0.103707
## `education.fctr Assoc-voc` -1.589 0.112020
## `education.fctr Prof-school` 1.954 0.050691 .
## `education.fctr 5th-6th` -1.492 0.135688
## `education.fctr 10th` -2.626 0.008635 **
## `education.fctr 1st-4th` 0.000 0.999676
## `education.fctr Preschool` 0.000 0.999833
## `education.fctr 12th` -0.007 0.994550
## age 3.507 0.000454 ***
## `race.fctr Black` 0.917 0.359078
## `race.fctr Asian-Pac-Islander` 0.925 0.354892
## `race.fctr Amer-Indian-Eskimo` 0.443 0.657450
## `race.fctr Other` 0.852 0.394420
## hoursperweek 2.828 0.004691 **
## `relationship.fctr Husband` 0.001 0.999507
## `relationship.fctr Wife` 0.001 0.999334
## `relationship.fctr Own-child` -0.898 0.369177
## `relationship.fctr Unmarried` 2.183 0.029005 *
## `relationship.fctr Other-relative` -0.006 0.995180
## `nativecountry.fctr Cuba` 1.908 0.056379 .
## `nativecountry.fctr Jamaica` 0.130 0.896757
## `nativecountry.fctr India` -0.843 0.399391
## `nativecountry.fctr Mexico` -0.008 0.993670
## `nativecountry.fctr South` 0.000 0.999976
## `nativecountry.fctr Puerto-Rico` -0.002 0.998153
## `nativecountry.fctr Honduras` NA NA
## `nativecountry.fctr England` 0.964 0.334943
## `nativecountry.fctr Canada` 1.056 0.291095
## `nativecountry.fctr Germany` -0.004 0.996508
## `nativecountry.fctr Iran` NA NA
## `nativecountry.fctr Philippines` -0.805 0.420791
## `nativecountry.fctr Italy` -0.001 0.998826
## `nativecountry.fctr Poland` -0.002 0.998740
## `nativecountry.fctr Columbia` -0.004 0.996746
## `nativecountry.fctr Cambodia` NA NA
## `nativecountry.fctr Thailand` NA NA
## `nativecountry.fctr Ecuador` -0.001 0.999169
## `nativecountry.fctr Laos` 0.258 0.796765
## `nativecountry.fctr Taiwan` 0.002 0.998006
## `nativecountry.fctr Haiti` -0.001 0.999054
## `nativecountry.fctr Portugal` -0.002 0.998770
## `nativecountry.fctr Dominican-Republic` NA NA
## `nativecountry.fctr El-Salvador` -0.005 0.996341
## `nativecountry.fctr France` -0.003 0.997383
## `nativecountry.fctr Guatemala` 0.001 0.998945
## `nativecountry.fctr China` -1.386 0.165625
## `nativecountry.fctr Japan` -0.002 0.998754
## `nativecountry.fctr Yugoslavia` NA NA
## `nativecountry.fctr Peru` -0.001 0.999048
## `nativecountry.fctr Outlying-US(Guam-USVI-etc)` NA NA
## `nativecountry.fctr Scotland` -0.002 0.998775
## `nativecountry.fctr Trinadad&Tobago` -0.002 0.998628
## `nativecountry.fctr Greece` -0.001 0.998907
## `nativecountry.fctr Nicaragua` -0.002 0.998705
## `nativecountry.fctr Vietnam` 0.000 0.999888
## `nativecountry.fctr Hong` -0.002 0.998585
## `nativecountry.fctr Ireland` -0.002 0.998572
## `nativecountry.fctr Hungary` NA NA
## `nativecountry.fctr Holand-Netherlands` NA NA
## `maritalstatus.fctr Married-civ-spouse` 0.000 0.999796
## `maritalstatus.fctr Divorced` 0.808 0.418880
## `maritalstatus.fctr Married-spouse-absent` -0.006 0.995434
## `maritalstatus.fctr Separated` 0.469 0.639391
## `maritalstatus.fctr Married-AF-spouse` -0.001 0.998927
## `maritalstatus.fctr Widowed` 1.244 0.213532
## `occupation.fctr Exec-managerial` 2.634 0.008448 **
## `occupation.fctr Handlers-cleaners` 0.033 0.973816
## `occupation.fctr Prof-specialty` 2.220 0.026412 *
## `occupation.fctr Other-service` -1.786 0.074026 .
## `occupation.fctr Sales` 2.045 0.040865 *
## `occupation.fctr Transport-moving` 0.241 0.809342
## `occupation.fctr Farming-fishing` -0.009 0.992542
## `occupation.fctr Machine-op-inspct` 0.592 0.553673
## `occupation.fctr Tech-support` 2.658 0.007851 **
## `occupation.fctr Craft-repair` 0.796 0.425807
## `occupation.fctr ?` 0.702 0.482461
## `occupation.fctr Protective-serv` -0.846 0.397550
## `occupation.fctr Armed-Forces` NA NA
## `occupation.fctr Priv-house-serv` -0.003 0.997300
## `workclass.fctr Self-emp-not-inc` 0.215 0.829567
## `workclass.fctr Private` 1.657 0.097516 .
## `workclass.fctr Federal-gov` 3.201 0.001370 **
## `workclass.fctr Local-gov` 0.446 0.655834
## `workclass.fctr ?` NA NA
## `workclass.fctr Self-emp-inc` 1.265 0.205836
## `workclass.fctr Without-pay` -0.002 0.998720
## `workclass.fctr Never-worked` NA NA
## capitalloss 0.931 0.351975
## `sex.fctr Female` -2.180 0.029227 *
## .rnorm -0.264 0.791562
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 1104.46 on 999 degrees of freedom
## Residual deviance: 545.16 on 913 degrees of freedom
## AIC: 719.16
##
## Number of Fisher Scoring iterations: 18## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading
## Reference
## Prediction N Y
## N 0 0
## Y 759 241
## over50k.fctr over50k.fctr.predict.Final.glm.N
## 1 N 0
## 2 Y 0
## over50k.fctr.predict.Final.glm.Y
## 1 759
## 2 241
## Reference
## Prediction N Y
## N 523 9
## Y 236 232
## over50k.fctr over50k.fctr.predict.Final.glm.N
## 1 N 523
## 2 Y 9
## over50k.fctr.predict.Final.glm.Y
## 1 236
## 2 232
## Reference
## Prediction N Y
## N 599 25
## Y 160 216
## over50k.fctr over50k.fctr.predict.Final.glm.N
## 1 N 599
## 2 Y 25
## over50k.fctr.predict.Final.glm.Y
## 1 160
## 2 216
## Reference
## Prediction N Y
## N 656 41
## Y 103 200
## over50k.fctr over50k.fctr.predict.Final.glm.N
## 1 N 656
## 2 Y 41
## over50k.fctr.predict.Final.glm.Y
## 1 103
## 2 200
## Reference
## Prediction N Y
## N 684 57
## Y 75 184
## over50k.fctr over50k.fctr.predict.Final.glm.N
## 1 N 684
## 2 Y 57
## over50k.fctr.predict.Final.glm.Y
## 1 75
## 2 184
## Reference
## Prediction N Y
## N 701 82
## Y 58 159
## over50k.fctr over50k.fctr.predict.Final.glm.N
## 1 N 701
## 2 Y 82
## over50k.fctr.predict.Final.glm.Y
## 1 58
## 2 159
## Reference
## Prediction N Y
## N 728 96
## Y 31 145
## over50k.fctr over50k.fctr.predict.Final.glm.N
## 1 N 728
## 2 Y 96
## over50k.fctr.predict.Final.glm.Y
## 1 31
## 2 145
## Reference
## Prediction N Y
## N 741 123
## Y 18 118
## over50k.fctr over50k.fctr.predict.Final.glm.N
## 1 N 741
## 2 Y 123
## over50k.fctr.predict.Final.glm.Y
## 1 18
## 2 118
## Reference
## Prediction N Y
## N 753 158
## Y 6 83
## over50k.fctr over50k.fctr.predict.Final.glm.N
## 1 N 753
## 2 Y 158
## over50k.fctr.predict.Final.glm.Y
## 1 6
## 2 83
## Reference
## Prediction N Y
## N 758 183
## Y 1 58
## over50k.fctr over50k.fctr.predict.Final.glm.N
## 1 N 758
## 2 Y 183
## over50k.fctr.predict.Final.glm.Y
## 1 1
## 2 58
## Reference
## Prediction N Y
## N 759 241
## Y 0 0
## over50k.fctr over50k.fctr.predict.Final.glm.N
## 1 N 759
## 2 Y 241
## over50k.fctr.predict.Final.glm.Y
## 1 0
## 2 0
## threshold f.score
## 1 0.0 0.3883965
## 2 0.1 0.6544429
## 3 0.2 0.7001621
## 4 0.3 0.7352941
## 5 0.4 0.7360000
## 6 0.5 0.6943231
## 7 0.6 0.6954436
## 8 0.7 0.6259947
## 9 0.8 0.5030303
## 10 0.9 0.3866667
## 11 1.0 0.0000000## [1] "Classifier Probability Threshold: 0.4000 to maximize f.score.fit"
## over50k.fctr over50k.fctr.predict.Final.glm.N
## 1 N 684
## 2 Y 57
## over50k.fctr.predict.Final.glm.Y
## 1 75
## 2 184
## Reference
## Prediction N Y
## N 684 57
## Y 75 184
## over50k.fctr over50k.fctr.predict.Final.glm.N
## 1 N 684
## 2 Y 57
## over50k.fctr.predict.Final.glm.Y
## 1 75
## 2 184
## Prediction
## Reference N Y
## N 684 75
## Y 57 184
## Accuracy Kappa AccuracyLower AccuracyUpper AccuracyNull
## 8.680000e-01 6.481520e-01 8.454346e-01 8.883720e-01 7.590000e-01
## AccuracyPValue McnemarPValue
## 5.968845e-18 1.389640e-01## Warning in mypredict_mdl(mdl, df = fit_df, rsp_var, rsp_var_out,
## model_id_method, : Expecting 1 metric: Accuracy; recd: Accuracy, Kappa;
## retaining Accuracy only
## model_id model_method
## 1 Final.glm glm
## feats
## 1 capitalgain, education.fctr, age, race.fctr, hoursperweek, relationship.fctr, nativecountry.fctr, maritalstatus.fctr, occupation.fctr, workclass.fctr, capitalloss, sex.fctr, .rnorm
## max.nTuningRuns min.elapsedtime.everything min.elapsedtime.final
## 1 1 2.535 0.502
## max.auc.fit opt.prob.threshold.fit max.f.score.fit max.Accuracy.fit
## 1 0.9325658 0.4 0.736 0.8120306
## max.AccuracyLower.fit max.AccuracyUpper.fit max.Kappa.fit min.aic.fit
## 1 0.8454346 0.888372 0.5100264 719.1617
## max.AccuracySD.fit max.KappaSD.fit
## 1 0.05885356 0.1288817glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="fit.data.training.all",
chunk_step_major=glb_script_df[nrow(glb_script_df), "chunk_step_major"],
chunk_step_minor=glb_script_df[nrow(glb_script_df), "chunk_step_minor"]+1,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed10 fit.data.training.all 6 0 110.406
## elapsed11 fit.data.training.all 6 1 117.134glb_rsp_var_out <- paste0(glb_rsp_var_out, tail(names(glb_models_lst), 1))
# Used again in predict.data.new chunk
glb_get_predictions <- function(df) {
if (glb_is_regression) {
df[, glb_rsp_var_out] <- predict(glb_fin_mdl, newdata=df, type="raw")
print(myplot_scatter(df, glb_rsp_var, glb_rsp_var_out,
smooth=TRUE))
df[, paste0(glb_rsp_var_out, ".err")] <-
abs(df[, glb_rsp_var_out] - df[, glb_rsp_var])
print(head(orderBy(reformulate(c("-", paste0(glb_rsp_var_out, ".err"))),
df)))
}
if (glb_is_classification && glb_is_binomial) {
# incorporate glb_clf_proba_threshold
# shd it only be for glb_fin_mdl or for earlier models ?
if ((prob_threshold <-
glb_models_df[glb_models_df$model_id == glb_fin_mdl_id, "opt.prob.threshold.fit"]) !=
glb_models_df[glb_models_df$model_id == glb_sel_mdl_id, "opt.prob.threshold.fit"])
stop("user specification for probability threshold required")
df[, paste0(glb_rsp_var_out, ".prob")] <-
predict(glb_fin_mdl, newdata=df, type="prob")[, 2]
df[, glb_rsp_var_out] <-
factor(levels(df[, glb_rsp_var])[
(df[, paste0(glb_rsp_var_out, ".prob")] >=
prob_threshold) * 1 + 1], levels(df[, glb_rsp_var]))
# prediction stats already reported by myfit_mdl ???
}
if (glb_is_classification && !glb_is_binomial) {
df[, glb_rsp_var_out] <- predict(glb_fin_mdl, newdata=df, type="raw")
}
return(df)
}
glb_trnent_df <- glb_get_predictions(glb_trnent_df)## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleadingprint(glb_feats_df <- mymerge_feats_importance(feats_df=glb_feats_df, sel_mdl=glb_fin_mdl,
entity_df=glb_trnent_df))## id cor.y exclude.as.feat cor.y.abs cor.low
## 3 capitalgain 0.23222327 0 0.23222327 1
## 5 education.fctr -0.06153968 0 0.06153968 1
## 2 age 0.24780278 0 0.24780278 1
## 10 race.fctr -0.01730292 0 0.01730292 1
## 6 hoursperweek 0.22402106 0 0.22402106 1
## 11 relationship.fctr -0.12845800 0 0.12845800 1
## 8 nativecountry.fctr -0.06238053 0 0.06238053 1
## 7 maritalstatus.fctr 0.01310876 0 0.01310876 1
## 9 occupation.fctr -0.12686813 0 0.12686813 1
## 13 workclass.fctr 0.07835390 0 0.07835390 1
## 4 capitalloss 0.10182189 0 0.10182189 1
## 12 sex.fctr -0.20283667 0 0.20283667 1
## 1 .rnorm -0.01983209 0 0.01983209 1
## importance
## 3 100.000000
## 5 77.459886
## 2 59.871191
## 10 54.648029
## 6 48.274918
## 11 45.387086
## 8 37.277684
## 7 37.226377
## 9 32.576969
## 13 21.237208
## 4 15.890756
## 12 15.794747
## 1 4.511675# Used again in predict.data.new chunk
glb_analytics_diag_plots <- function(obs_df) {
for (var in subset(glb_feats_df, !is.na(importance))$id) {
plot_df <- melt(obs_df, id.vars=var,
measure.vars=c(glb_rsp_var, glb_rsp_var_out))
# if (var == "<feat_name>") print(myplot_scatter(plot_df, var, "value",
# facet_colcol_name="variable") +
# geom_vline(xintercept=<divider_val>, linetype="dotted")) else
print(myplot_scatter(plot_df, var, "value", colorcol_name="variable",
facet_colcol_name="variable", jitter=TRUE) +
guides(color=FALSE))
}
if (glb_is_regression) {
plot_vars_df <- subset(glb_feats_df, importance > glb_feats_df[glb_feats_df$id == ".rnorm", "importance"])
print(myplot_prediction_regression(df=obs_df,
feat_x=ifelse(nrow(plot_vars_df) > 1, plot_vars_df$id[2], ".rownames"),
feat_y=plot_vars_df$id[1],
rsp_var=glb_rsp_var, rsp_var_out=glb_rsp_var_out,
id_vars=glb_id_vars)
# + facet_wrap(reformulate(plot_vars_df$id[2])) # if [1 or 2] is a factor
# + geom_point(aes_string(color="<col_name>.fctr")) # to color the plot
)
}
if (glb_is_classification) {
if (nrow(plot_vars_df <- subset(glb_feats_df, !is.na(importance))) == 0)
warning("No features in selected model are statistically important")
else print(myplot_prediction_classification(df=obs_df,
feat_x=ifelse(nrow(plot_vars_df) > 1, plot_vars_df$id[2],
".rownames"),
feat_y=plot_vars_df$id[1],
rsp_var=glb_rsp_var,
rsp_var_out=glb_rsp_var_out,
id_vars=glb_id_vars)
# + geom_hline(yintercept=<divider_val>, linetype = "dotted")
)
}
}
glb_analytics_diag_plots(obs_df=glb_trnent_df)
## age workclass education maritalstatus
## 6 37 Private Masters Married-civ-spouse
## 355 43 Private Bachelors Married-civ-spouse
## 418 36 Federal-gov Masters Never-married
## 1250 51 Local-gov Bachelors Married-civ-spouse
## 1806 38 Private Bachelors Married-civ-spouse
## 2388 61 Self-emp-not-inc HS-grad Married-civ-spouse
## 5367 44 Private Prof-school Married-civ-spouse
## occupation relationship race sex capitalgain
## 6 Exec-managerial Wife White Female 0
## 355 Transport-moving Husband White Male 0
## 418 Prof-specialty Not-in-family White Male 0
## 1250 Adm-clerical Husband White Male 0
## 1806 Craft-repair Wife White Female 0
## 2388 Craft-repair Husband White Male 0
## 5367 Prof-specialty Husband White Male 99999
## capitalloss hoursperweek nativecountry over50k workclass.fctr
## 6 0 40 United-States <=50K Private
## 355 0 45 United-States <=50K Private
## 418 1408 40 United-States <=50K Federal-gov
## 1250 0 40 United-States >50K Local-gov
## 1806 0 40 United-States <=50K Private
## 2388 0 52 United-States >50K Self-emp-not-inc
## 5367 0 65 United-States >50K Private
## education.fctr maritalstatus.fctr occupation.fctr
## 6 Masters Married-civ-spouse Exec-managerial
## 355 Bachelors Married-civ-spouse Transport-moving
## 418 Masters Never-married Prof-specialty
## 1250 Bachelors Married-civ-spouse Adm-clerical
## 1806 Bachelors Married-civ-spouse Craft-repair
## 2388 HS-grad Married-civ-spouse Craft-repair
## 5367 Prof-school Married-civ-spouse Prof-specialty
## relationship.fctr race.fctr sex.fctr nativecountry.fctr .rnorm
## 6 Wife White Female United-States 1.5640010
## 355 Husband White Male United-States -1.5273574
## 418 Not-in-family White Male United-States -0.2095156
## 1250 Husband White Male United-States -0.5379833
## 1806 Wife White Female United-States 1.1681171
## 2388 Husband White Male United-States -1.5651562
## 5367 Husband White Male United-States -0.8259619
## over50k.fctr over50k.fctr.predict.Final.glm.prob
## 6 N 0.7861746
## 355 N 0.5357181
## 418 N 0.5211014
## 1250 Y 0.3400039
## 1806 N 0.6203245
## 2388 Y 0.2277406
## 5367 Y 1.0000000
## over50k.fctr.predict.Final.glm
## 6 Y
## 355 Y
## 418 Y
## 1250 N
## 1806 Y
## 2388 N
## 5367 Y
## over50k.fctr.predict.Final.glm.accurate .label
## 6 FALSE .6
## 355 FALSE .355
## 418 FALSE .418
## 1250 FALSE .1250
## 1806 FALSE .1806
## 2388 FALSE .2388
## 5367 TRUE .5367
replay.petrisim(pn=glb_analytics_pn,
replay.trans=(glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
"data.training.all.prediction","model.final")), flip_coord=TRUE)## time trans "bgn " "fit.data.training.all " "predict.data.new " "end "
## 0.0000 multiple enabled transitions: data.training.all data.new model.selected firing: data.training.all
## 1.0000 1 2 1 0 0
## 1.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction firing: data.new
## 2.0000 2 1 1 1 0
## 2.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction data.new.prediction firing: model.selected
## 3.0000 3 0 2 1 0
## 3.0000 multiple enabled transitions: model.final data.training.all.prediction data.new.prediction firing: data.training.all.prediction
## 4.0000 5 0 1 1 1
## 4.0000 multiple enabled transitions: model.final data.training.all.prediction data.new.prediction firing: model.final
## 5.0000 4 0 0 2 1
glb_script_df <- rbind(glb_script_df,
data.frame(chunk_label="predict.data.new",
chunk_step_major=max(glb_script_df$chunk_step_major)+1,
chunk_step_minor=0,
elapsed=(proc.time() - glb_script_tm)["elapsed"]))
print(tail(glb_script_df, 2))## chunk_label chunk_step_major chunk_step_minor elapsed
## elapsed11 fit.data.training.all 6 1 117.134
## elapsed12 predict.data.new 7 0 127.038Step 7: predict data.new
# Compute final model predictions
glb_newent_df <- glb_get_predictions(glb_newent_df)## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleading## Warning in predict.lm(object, newdata, se.fit, scale = 1, type =
## ifelse(type == : prediction from a rank-deficient fit may be misleadingglb_analytics_diag_plots(obs_df=glb_newent_df)
## age workclass education maritalstatus
## 31 20 Private Some-college Never-married
## 117 40 Private Bachelors Married-civ-spouse
## 608 52 Self-emp-inc Bachelors Married-civ-spouse
## 810 57 Private HS-grad Widowed
## 1311 40 Private HS-grad Married-civ-spouse
## 1462 32 Private Some-college Married-civ-spouse
## 1468 32 Private 11th Divorced
## 23048 42 Self-emp-not-inc Masters Married-civ-spouse
## occupation relationship race sex capitalgain
## 31 Sales Own-child Black Male 0
## 117 Exec-managerial Husband White Male 0
## 608 Sales Husband White Male 0
## 810 Transport-moving Unmarried White Male 0
## 1311 Exec-managerial Husband White Male 0
## 1462 Sales Husband White Male 0
## 1468 Sales Not-in-family White Male 0
## 23048 Sales Husband White Male 99999
## capitalloss hoursperweek nativecountry over50k workclass.fctr
## 31 0 44 United-States <=50K Private
## 117 0 60 United-States <=50K Private
## 608 0 55 United-States <=50K Self-emp-inc
## 810 653 42 United-States >50K Private
## 1311 0 40 United-States >50K Private
## 1462 0 48 United-States <=50K Private
## 1468 0 43 United-States >50K Private
## 23048 0 80 United-States >50K Self-emp-not-inc
## education.fctr maritalstatus.fctr occupation.fctr
## 31 Some-college Never-married Sales
## 117 Bachelors Married-civ-spouse Exec-managerial
## 608 Bachelors Married-civ-spouse Sales
## 810 HS-grad Widowed Transport-moving
## 1311 HS-grad Married-civ-spouse Exec-managerial
## 1462 Some-college Married-civ-spouse Sales
## 1468 11th Divorced Sales
## 23048 Masters Married-civ-spouse Sales
## relationship.fctr race.fctr sex.fctr nativecountry.fctr .rnorm
## 31 Own-child Black Male United-States -0.08837181
## 117 Husband White Male United-States -3.11383046
## 608 Husband White Male United-States 0.62036514
## 810 Unmarried White Male United-States -1.73102184
## 1311 Husband White Male United-States 1.82063036
## 1462 Husband White Male United-States 0.44603769
## 1468 Not-in-family White Male United-States 0.13299383
## 23048 Husband White Male United-States -0.78729784
## over50k.fctr over50k.fctr.predict.Final.glm.prob
## 31 N 1.977742e-02
## 117 N 8.424240e-01
## 608 N 8.252479e-01
## 810 Y 2.238346e-01
## 1311 Y 3.477002e-01
## 1462 N 4.564036e-01
## 1468 Y 2.486192e-09
## 23048 Y 1.000000e+00
## over50k.fctr.predict.Final.glm
## 31 N
## 117 Y
## 608 Y
## 810 N
## 1311 N
## 1462 Y
## 1468 N
## 23048 Y
## over50k.fctr.predict.Final.glm.accurate .label
## 31 TRUE .31
## 117 FALSE .117
## 608 FALSE .608
## 810 FALSE .810
## 1311 FALSE .1311
## 1462 FALSE .1462
## 1468 FALSE .1468
## 23048 TRUE .23048
tmp_replay_lst <- replay.petrisim(pn=glb_analytics_pn,
replay.trans=(glb_analytics_avl_objs <- c(glb_analytics_avl_objs,
"data.new.prediction")), flip_coord=TRUE)## time trans "bgn " "fit.data.training.all " "predict.data.new " "end "
## 0.0000 multiple enabled transitions: data.training.all data.new model.selected firing: data.training.all
## 1.0000 1 2 1 0 0
## 1.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction firing: data.new
## 2.0000 2 1 1 1 0
## 2.0000 multiple enabled transitions: data.training.all data.new model.selected model.final data.training.all.prediction data.new.prediction firing: model.selected
## 3.0000 3 0 2 1 0
## 3.0000 multiple enabled transitions: model.final data.training.all.prediction data.new.prediction firing: data.training.all.prediction
## 4.0000 5 0 1 1 1
## 4.0000 multiple enabled transitions: model.final data.training.all.prediction data.new.prediction firing: model.final
## 5.0000 4 0 0 2 1
## 6.0000 6 0 0 1 2
print(ggplot.petrinet(tmp_replay_lst[["pn"]]) + coord_flip())
Null Hypothesis (\(\sf{H_{0}}\)): mpg is not impacted by am_fctr.
The variance by am_fctr appears to be independent. #{r q1, cache=FALSE} # print(t.test(subset(cars_df, am_fctr == "automatic")$mpg, # subset(cars_df, am_fctr == "manual")$mpg, # var.equal=FALSE)$conf) # We reject the null hypothesis i.e. we have evidence to conclude that am_fctr impacts mpg (95% confidence). Manual transmission is better for miles per gallon versus automatic transmission.
## chunk_label chunk_step_major chunk_step_minor elapsed
## 10 fit.models 5 1 98.841
## 11 fit.data.training.all 6 0 110.406
## 13 predict.data.new 7 0 127.038
## 12 fit.data.training.all 6 1 117.134
## 6 extract_features 3 0 6.361
## 2 cleanse_data 2 0 1.654
## 7 select_features 4 0 7.735
## 4 manage_missing_data 2 2 2.860
## 5 encode_retype_data 2 3 3.141
## 8 remove_correlated_features 4 1 7.936
## 9 fit.models 5 0 7.989
## 3 inspectORexplore.data 2 1 1.692
## 1 import_data 1 0 0.002
## elapsed_diff
## 10 90.852
## 11 11.565
## 13 9.904
## 12 6.728
## 6 3.220
## 2 1.652
## 7 1.374
## 4 1.168
## 5 0.281
## 8 0.201
## 9 0.053
## 3 0.038
## 1 0.000## [1] "Total Elapsed Time: 127.038 secs"
## R version 3.1.3 (2015-03-09)
## Platform: x86_64-apple-darwin13.4.0 (64-bit)
## Running under: OS X 10.10.3 (Yosemite)
##
## locale:
## [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
##
## attached base packages:
## [1] tcltk grid stats graphics grDevices utils datasets
## [8] methods base
##
## other attached packages:
## [1] randomForest_4.6-10 rpart.plot_1.5.2 rpart_4.1-9
## [4] ROCR_1.0-7 gplots_2.16.0 caret_6.0-41
## [7] lattice_0.20-31 sqldf_0.4-10 RSQLite_1.0.0
## [10] DBI_0.3.1 gsubfn_0.6-6 proto_0.3-10
## [13] reshape2_1.4.1 plyr_1.8.1 caTools_1.17.1
## [16] doBy_4.5-13 survival_2.38-1 ggplot2_1.0.1
##
## loaded via a namespace (and not attached):
## [1] bitops_1.0-6 BradleyTerry2_1.0-6 brglm_0.5-9
## [4] car_2.0-25 chron_2.3-45 class_7.3-12
## [7] codetools_0.2-11 colorspace_1.2-6 compiler_3.1.3
## [10] digest_0.6.8 e1071_1.6-4 evaluate_0.5.5
## [13] foreach_1.4.2 formatR_1.1 gdata_2.13.3
## [16] gtable_0.1.2 gtools_3.4.1 htmltools_0.2.6
## [19] iterators_1.0.7 KernSmooth_2.23-14 knitr_1.9
## [22] labeling_0.3 lme4_1.1-7 MASS_7.3-40
## [25] Matrix_1.2-0 mgcv_1.8-6 minqa_1.2.4
## [28] munsell_0.4.2 nlme_3.1-120 nloptr_1.0.4
## [31] nnet_7.3-9 parallel_3.1.3 pbkrtest_0.4-2
## [34] quantreg_5.11 RColorBrewer_1.1-2 Rcpp_0.11.5
## [37] rmarkdown_0.5.1 scales_0.2.4 SparseM_1.6
## [40] splines_3.1.3 stringr_0.6.2 tools_3.1.3
## [43] yaml_2.1.13